(Kenneth M. Goldstein) Interest Groups, Lobbying, (B-Ok - Xyz)

Interest Groups, Lobbying, and
Participation in America
Understanding why individuals participate in politics demands attention to
more than just individual attributes and attitudes. Similarly, understanding
how interest groups influence policy making demands attention to more
than just the financial donations and direct activities of Washington-based
lobbyists. To answer fundamental questions about what determines when
and why people participate in politics and how organized interests go about
trying to influence legislative decision making, we must understand how and
why political leaders recruit which members of the public into the political
arena. Looking from the bottom up with survey data and from the top down
with data from interest group interviews, Kenneth Goldstein develops and
tests a theory of how tactical choices in a grass roots campaign are made. In
doing so, he demonstrates that outside lobbying activities deserve a place in
any correctly specified model of interest group influence, political participa-
tion, or legislative decision making.
Kenneth M. Goldstein is assistant professor of political science at Arizona

State University. He earned his Ph.D. in political science at the University of
Michigan. His published work has appeared in the American Journal of
Political Science and Public Opinion Quarterly and in a series of book chapters.
He is currently at work on a project on television advertising. Goldstein has
also had extensive professional experience in conducting survey research for
media and political clients. He currently works as a consultant for the ABC
News political unit.
Interest Groups,
Lobbying, and
Participation in
America
KENNETH M. GOLDSTEIN
Arizona State University
CAMBRIDGE
UNIVERSITY PRESS
PUBLISHED BY THE PRESS SYNDICATE OF THE UNIVERSITY OF CAMBRIDGE
The Pitt Building, Trumpington Street, Cambridge, United Kingdom
CAMBRIDGE UNIVERSITY PRESS
The Edinburgh Building, Cambridge CB2 2RU, UK www.cup.cam.ac.uk
40 West 20th Street, New York, NY 10011-4211, USA www.cup.org
10 Stamford Road, Oakleigh, Melbourne 3166, Australia
Ruiz de Alarcon 13, 28014 Madrid, Spain
© Kenneth M. Goldstein 1999
This book is in copyright. Subject to statutory exception

and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without
the written permission of Cambridge University Press.
First published 1999

Typeface Plantin 10/12 System MagnaType™ [AG]
A catalog record for this book is available from

the British Library.
Library of Congress Cataloging in Publication data

Goldstein, Kenneth M., 1965-
Interest groups, lobbying, and participation in America / Kenneth
M. Goldstein,
p. cm.
Includes bibliographical references.
ISBN 0-521-63047-9. - ISBN 0-521-63962-X (pbk.)
1. Pressure groups - United States. 2. Lobbying - United States.
3. Political participation - United States. I. Title.
JK1118.G634 1999
324'.4'09732 - dc21 98-53581
CIP
ISBN 0 521 63047 9 hardback
ISBN 0 521 63962 X paperback
Transferred to digital printing 2003

For Dan
Contents
List of Figures and Tables page viii

Acknowledgments xi
1 Introduction 1
2 Patterns and Puzzles in Participation and Lobbying 14
3 The Political Logic of Political Decisions 30
4 Explaining Lobbying Decisions 53
5 Lobbying Decisions and the Health Care Reform Battle 72
6 Patterns of Recruitment and Participation in the Mass Public 106
7 Conclusion 125
Appendix A: Sources and Coding for Survey Data 133

Appendix B: Interest Group Sampling Frame 138
Appendix C: Chronology of Health Care Reform Legislation 144
Bibliography 147
Index 153
vn
Figures and Tables
Figures
2.1 Mail to House of Representatives, 1973-1994 page 16
2.2 Percentage of citizens contacting Congress to express
opinion on issues, 1978-1994 17
2.3 Grass roots lobbying mentions in the press, 1977-1994 25
3.1 Major factors influencing congressional elections 32
3.2 Legislative decision making 37
6.1 Mobilization and participation in target and nontarget
states 111
Tables
1.1 Issue areas and frequencies 7
1.2 Sample characteristics and response rates 9
3.1 Strategies for influencing congressional elections 35
3.2 Strategies to influence congressional decisions 42
3.3 Objectives, strategies, and legislative targets 48
4.1 Strategic objective by number of mobilization campaigns 55
4.2 Timing of mobilization campaigns with a legislative
objective 59
4.3 Member targets with a legislative objective 60
4.4 Constituent mobilization targets with a legislative
objective 62
4.5 Constituent mobilization targets with a legislative
objective 64
4.6 Member targets with an electoral objective 68
4.7 Constituent mobilization targets with an electoral
objective 69
viii
List of Figures and Tables ix
4.8 Constituent mobilization targets with an electoral
objective 70
5.1 Interviews by type of group 75
5.2 Where and when groups mobilized communications 78
5.3 Measures of committee competitiveness 80
5.4 Committee targets 81
5.5 Legislative targets within committee 84
5.6 Constituent mobilization targets and levels of leverage 87
5.7 Constituent mobilization targets 88
5.8 Handicapping House Energy and Commerce
Committee 92
5.9 Member targets in House Energy and Commerce
Committee 94
5.10 Handicapping House Ways and Means Committee 98
5.11 Member targets in House Ways and Means Committee 100
5.12 Handicapping Senate Finance Committee 102
5.13 Member targets in Senate Finance Committee 103
6.1 Targeted states 108
6.2 Personal characteristics and communicating with
Congress on health care 110
6.3 Personal characteristics and being mobilized to contact
Congress 112
6.4 Causes of phoning, faxing, or writing Congress on
healthcare 113
6.5 Causes of being mobilized 114
6.6 Representation ratios for being mobilized 116
6.7 Targeted House districts 118
6.8 Individual characteristics and communicating to
Congress 119
6.9 Causes of contacting Congress 120
6.10 Contacting Congress and partisanship 122
6.11 Proportion of Party identifiers and party voters by time
of contact 122
6.12 Communicating with Congress to express opinion on
issues, 1980-1994 123
Acknowledgments
As I finish this project, it's hard to know where to start the acknowledg-
ments. I first owe a debt to the representatives of the many interest groups
with whom I spoke. Without their time and openness, I would not have been
able to complete this project. I would also like to thank the numerous
journalists and consultants with whom I spoke.
I owe a great debt to my teachers and dissertation committee members at
Michigan. Steve Rosenstone's advice and previous work were of great help
in this project. Mayer Zald, from the sociology department, introduced me
to a new literature and pushed me to think about my project from different
angles. In fact, my dissertation grew out of a paper I wrote in his seminar on
social movements. I thoroughly enjoyed the hours we talked about social
movements, interest groups, and Michigan sports. Chris Achen and Greg
Markus were also great teachers and friends. They both constantly pushed
me to be rigorous in my theory and methods, but also understood my
passion for politics and were fellow "political junkies." They gave early
drafts careful reads. I enjoyed their insight and company tremendously dur-
ing my time in Ann Arbor.
Although he was not on my committee, I would also like to thank John
Kingdon. His research strategies and advice clearly influenced my work.
Roy Pierce was a teacher and valued friend in Ann Arbor. Arlene Saxon-
house gave me the chance to teach a class in Washington, D.C., which gave
me the opportunity to conduct the preliminary fieldwork for this project. Zvi
and Marlene Gitelman treated me like family during my time in Ann Arbor.
In 1995 I was the recipient of a Gerald R. Ford Dissertation Fellowship.
This award enabled me to conduct the large volume of interest group inter-
views. The Ford Fellowships are a tremendous resource that the department
offers and I am very grateful.
Since I received help from so many people, I hesitate to single out one
individual. I must note, however, the special influence Kent Jennings has
had on my education over the past six years. From the class I took with Kent
xi
xii Acknowledgments
the first semester of my first year of graduate school to his work as cochair of
my dissertation committee, he has been an inspirational teacher, mentor,
role model, and friend. He is simply a class act, and I owe him a special debt
of gratitude.
There are not words that can express my appreciation to my parents for all
that they have done. They have provided love and support and I cannot
thank them enough. Finally, I would like to thank my wife Amanda. She
started this project as a girlfriend and she ends it as my wife and best friend.
Her patience, love, and support - not to mention her careful editing - made
this project possible.
1
Introduction
Grass roots 1. The common people.

2. The basic source or support.
Vfebster's New World Dictionary
Grass roots: The ultmate source of power, usually

patronized, occasionally feared.
Safire 's Political Dictionary
In the summer of 1982, Senator Bob Dole, then chairman of the Senate
Finance Committee, and Representative Dan Rostenkowski, then chairman
of the House Ways and Means Committee, introduced legislation to with-
hold taxes on interest from bank accounts and dividends from securities.
Proponents of the bill argued that most other types of income were already
subject to withholding and that this legislation would simply plug a major
tax loophole and tap a notorious source of unreported income. Arguments
of this kind apparently convinced large majorities of both houses of Con-
gress, and just before the August recess, the bill comfortably passed the
Republican-controlled Senate and the Democrat-controlled House. The
bill was signed by President Reagan and was scheduled to take effect within
the year.
Fearful, however, of the multibillion-dollar cost of enforcing the law, the
banking industry dropped "the hydrogen bomb of modern day lobbying, an
effort whose firepower was awesome, whose carnage was staggering. In one
fell swoop down went the chairman of the Senate Finance Committee, down
went the chairman of the House Ways and Means Committee, down went
the Secretary of the Treasury, down went the president of the United States"
(Taylor 1983, A12). Led by the American Bankers Association (ABA) and
the U.S. League of Savings, the banking industry used newspaper advertise-
ments, posters in branches, and, most importantly, inserts in the monthly
statements typically sent to all customers to encourage people to contact
1
2 Interest Groups, Lobbying, and Participation
Congress in opposition to the new law. The effort, orchestrated by the

Chicago advertising and public relations firm of Leo Burnett and Company,
deluged Congress with more than twenty-two million constituent com-
munications (Berry 1989; Taylor 1983; Wittenberg and Wittenberg 1994;
Wolpe 1990). Weeks later the House (382 to 41) and the Senate (94 to 5)
reversed themselves and overwhelmingly repealed the withholding on inter-
est and dividend income earned by individuals.
In a similar vein, Catastrophic Care legislation in 1989 began with wide
bipartisan support and ended up being overwhelmingly repealed. The legis-
lation was introduced as a way to protect elderly and disabled Americans
from huge hospital and doctor bills. It resulted in a powerful Democratic
committee chairman literally being chased through the halls of Congress by
angry gangs of elderly constituents. Again, the goal of the legislation was to
protect elderly and disabled Americans from astronomical medical bills.
The mechanism to finance this benefit was a supplemental Medicare pre-
mium capped at $800. Although 40 percent of seniors would have been
required to pay an extra premium, only about one in twenty elderly Ameri-
cans would have had to pay the full capped amount (Wolpe 1990).
Groups like James Roosevelt's (FDR's oldest son) National Committee to
Preserve Social Security and Medicare, however, opposed any increase in
premiums no matter what the new benefits were. The group sent out mail-
ings telling itsfivemillion members that the new law was a "seniors only tax"
and suggested that all seniors would be forced to pay an extra tax for benefits
that they already had. "Your Federal Taxes for 1989 may increase by up to
$1,600 just because you are over the age of 65!" (Hosenball 1989). The
claims of Roosevelt's group and other direct-mail organizations were half-
truths at best. Still, millions of elderly Americans contacted their members
of Congress and the bill was repealed less than a year later.
Now consider the dilemma of the "Big Three" automakers during the
1990 debate over the Clean-Air Act. Newsweek magazine (1991) described
the challenge facing automobile companies: "How could they [the auto-
makers] squash legislation that improved fuel efficiency, reduced air pollu-
tion and reduced dependence on foreign oil without looking like greedy
corporate ghouls?" Jack Bonner, a prominent grass roots consultant, rea-
soned that smaller cars would hurt the elderly, the disabled, and those who
must transport children.1 So, in a matter of days, Bonner's "shock yuppies"
1
Jack Bonner, president of Bonner & Associates, is probably the most well known grass roots
consultant in Washington. In fact, it is nearly impossible to read a newspaper article or have a
conversation about grass roots politics in Washington without Bonner's name popping up.
Using more than two hundred articulate "unemployed policy junkies," Bonner's firm will
scour congressional districts for groups and individuals to contact their representative from
the grass roots in support of or in opposition to legislation of concern to his clients (Brinkley
1993b; Browning 1994; Gugliotta 1994). There is nothing particularly complicated about
what Bonner does and he does not attempt to hide his efforts. He warns a Hill office when he
Introduction 3
contacted elderly organizations, disabled groups, and the Boy Scouts in the
constituencies of key conference committee members and created a torrent
of opposition to higher fuel standards. In this way, Bonner helped change
what easily could have been framed as an antienvironment vote into a pro-
elderly, pro-disabled, and pro-Boy Scouts vote {Newsweek 1991).
In 1991, the American Bankers Association also turned to Bonner for
help. In the fall of that year, the Senate had passed an amendment that
would have regulated the interest rates banks could charge on credit cards.
Because millions of Americans carried monthly balances on their credit card
accounts, the bill looked like a winner. It was also one that would have taken
a huge bite out of a major source of banking profits. In a four-day period,
Bonner's firm was able to generate ten thousand phone calls from voters and
community leaders in ten districts represented by members of the House
Banking, Finance, and Urban Affairs Committee (Stone 1993, 755). When
all was said and done, the amendment was not included in the House
version of the bill and was subsequently dropped in a House and Senate
conference committee.
The preceding cases provide prominent examples of what political scien-
tists categorize as outside lobbying and what is known on Capitol Hill and in
the professional community as grass roots lobbying or issue advocacy cam-
paigns. Although there is no real formal definition of the tactic, popular and
scholarly accounts alike consider grass roots or outside lobbying to be any
type of action that attempts to influence inside-the-beltway inhabitants by
influencing the attitudes or behavior of outside-the-beltway inhabitants. It
stands in contrast to "inside strategies" such as private meetings with mem-
bers and staff, testifying at committee hearings, and contributing money.
Grass roots lobbying is akin to the "going public" strategy "whereby a
president promotes himself and his policies in Washington by appealing to
the American public for support" (Kernell 1993, 2). Perhaps the best defini-
tion comes from a trade association executive speaking at a workshop on
grass roots lobbying. He defined grass roots lobbying as "The identification,
recruitment, and mobilization of constituent-based political strength capa-
ble of influencing political decisions."
Although it is tempting to make judgments about the effect of such inter-
est group tactics, my goal in presenting these four examples was not to prove
that grass roots lobbying was the decisive factor in the legislative battles. The
four were complicated issues, and there are multiple explanations for the
eventual outcomes in each case. For example, the Clean Air Act had been
debated for years (Cohen 1992), and a stock market drop and the comments
is about to mobilize citizens in its district so that it can allocate staff time in advance to
process the incoming faxes, letters, and phone calls. In fact, Bonner is even known to send
flowers and chocolates to Hill receptionists the day before his efforts hit.
of legions of financial experts preceded the deletion of caps on credit card

interest in the final banking bill (Knight 1991). Besides, in many other
instances grass roots lobbying had no apparent effect.
These four cases do, however, illustrate how grass roots lobbying can be
an effective tool for lobbyists to convey information. More specifically, these
cases illustrate how grass roots lobbying can signal legislators on the elec-
toral consequences of their actions and provide information to constituents
that may reframe an issue and possibly change mass opinion. These four
cases suggest that understanding why and how lobbying choices are made is
a crucial first step toward understanding both the character of mass par-
ticipation and the nature of interest group influence. With our current level
of understanding, the appropriate question is not whether orchestrated
communications have no effect (as many political scientists have assumed)
or whether they decide every issue (as many journalists have assumed), but
instead, why and how they are used? At the very least, these cases - and the
recent explosion in participation and mobilization that I will describe in the
next chapter - strongly suggest that there is probable cause to investigate
when, where, why, and how interest groups go public.
The Argument
Cases such as the ones just described, as well as the apparent growth in both
communications to Congress and the use of grass roots campaigns as a
lobbying tactic, were the inspiration for this book. Yet my brief description
of these four grass roots campaigns, which attempted to stimulate constitu-
ent communications, really provide no startling new information and would
not surprise scholarly observers of interest groups or mass participation.
After all, scholars of participation are well aware of the impact of elite
behavior in stimulating various types of mass participation (Rosenstone and
Hansen 1993; Verba, Schlozman, and Brady 1995a). Likewise, students of
lobbying are well aware that outside tactics such as stimulating constituent
communications to Congress are an important weapon in an interest
group's arsenal of tactics (Schlozman and Tierney 1986).
The central argument made in this book is that the elite stimulus of mass
participation binds together crucial questions of group influence and indi-
vidual participation: When and why do people participate in politics? How
do organized interests decide when, where, and how to influence public
policy? These questions about mass participation and group influence, usu-
ally tackled separately by political scientists, strike at the heart of arguments
about the state of democracy in America. To answer these two fundamental
questions - to understand what determines when and why people partici-
pate in politics and how organized interests go about trying to influence
Introduction 5
legislative decision making - we must understand how and why political
leaders recruit which members of the public into the political arena.
My goal is to develop and test a theory of how tactical choices in a grass
roots campaign are made. My goal is to develop and test a theory of how
tactical choices about when to lobby, where to lobby, and whom to lobby are
made. In doing so, I demonstrate that outside lobbying activities deserve a
place in any correctly specified model of interest group influence, political
participation, or legislative decision making. Put another way, an under-
standing of why individuals participate in politics demands attention to
more than just individual attributes and attitudes; and an understanding of
how interest groups influence policy making demands attention to more
than just the financial donations and direct activities of Washington-based
lobbyists.
Yet, understanding how and why lobbying choices are made is difficult if
we do not have an accurate theoretical picture of the strategic logic govern-
ing the use of grass roots lobbying. Although scholars have painted a more
detailed theoretical and empirical picture of interest group formation and
behavior in the past decade, the motives and goals of lobbyists are still not
thoroughly understood. In fact, not knowing where to look may be one of
the primary reasons why scholars have found it difficult to demonstrate
interest group influence. For example, one of the problems with previous
work on interest groups and lobbying has been that scholars have tried to
identify the direct independent influence of particular strategies and tactics.
A typical question is. With all else held constant, what is the independent
direct influence of a PAC donation on a roll call vote? As these cases suggest,
however, lobbyists may work indirectly through constituents to influence
congressional decisions. A more detailed understanding of the strategic
objectives of lobbying is required.
Similarly, in the mass participation literature, attention to the effects or
goals of citizen activity is often divorced from attention to the causes or
correlates of citizen activity. Indeed, the standard style in many studies of
participation is to stipulate that participation matters and then to identify its
causes. I contend that who participates and how participation matters are
questions that should not be studied separately. It is not possible to under-
stand elite efforts to stimulate mass participation and communications to
Congress without also understanding the politics and political context in
which grass roots lobbying and communications to Congress take place.
Data and Methodology

This book reports on empirical research designed to understand how and
why interest groups utilize outside lobbying tactics. Although this study is
primarily concerned with why and how interest groups decide which citizens
to recruit into the political arena, I also discuss a broader range of tactics
that interest groups use when they go public. To investigate the strategic
calculations and decisions made by organized interests in such grass roots or
outside lobbying campaigns, I utilize multiple sources of data collected in
different ways and at different times.
One set of data utilized in this book consists of public opinion surveys.
During the summer of 1994,1 was able to add participation and recruitment
questions to the Battleground Poll, a national survey jointly conducted by a
Democratic polling firm and a Republican polling firm.2 Although many
surveys contain "contact Congress" questions and many surveys contain an
"electoral mobilization" question, few studies specifically ask whether a
respondent was recruited to lobby Congress and with respect to what issue.
Data from the Battleground Poll enable me to test specifically the effect of
recruitment contacts on individuals and to investigate who, in fact, is being
recruited.
In addition, a large-sample Times-Mirror survey conducted in July 1994
provides another barometer of citizen participation during the identical time
period and also affords another opportunity to investigate the effect of tar-
geted districts on political participation. I also use data from the National
Election Studies (NES) in this book. The NES data provide a time series on
participation, permitting me to analyze how the quantity and partisan com-
position of those who contact Congress have changed over time.
These data sources are certainly valuable for my purposes. Nevertheless,
to understand grass roots lobbying and mass participation, it is vital to go
beyond the sample survey. Accordingly, the main body of research in this
book comes from data gathered in forty-one in-depth personal interviews
with interest group representatives.3 The unit of analysis, however, is nei-
ther the individual whom I interviewed nor the group he or she represented.
It is instead the individual grass roots lobbying campaigns and tactical
choices that were made. In most of the interviews, more than one lobbying
campaign and more than one set of tactical choices were discussed. Alto-
gether, the forty-one interviews covered ninety-four separate grass roots
lobbying campaigns across fifteen issue domains. Table 1.1 lists the issues
discussed and their frequency of occurrence.
2
Thanks to Celinda Lake, now of Lake Research, Ed Goeas of The Tarrance Group, and Lori
Gudermuth, now of Public Opinion Strategies, for allowing me to insert recruitment and
participation questions in the Battleground Poll.
3
I did not enjoy a good response from a pretest of interest groups that I conducted by mail. On
the other hand, in pretest interviews that I conducted in person, I was able to gather both
information about specific mobilization decisions as well as in-depth background informa-
tion on how grass roots mobilization tactics were utilized.
Introduction
Table 1.1. Issue Areas and Frequencies
Number of Mobilization
Issue Efforts Examined
Health Care 21
Clinton Budget/Stimulus Package 12
NAFTA 12
Medicare * 10
Crime Bill/Assault Weapons 8
Balanced Budget Amendment 6
Federal Funding Abortion 5
Tort Reform * 4
Term Limits * 4
Telecommunications * 4
Campaign Finance 2
Lobbying Reform 2
Smoking/Tobacco 2
Worker Safety * 1
Meat Inspections * 1
Total 94
* 104th Congress.
Source: Author's interviews.
The forty-one interviews come from a sample of eighty organizations that

were drawn from a list I compiled of 191 interest groups pursuing grass
roots tactics in the 103rd Congress. Compiling such a sampling frame was
not a straightforward task. Although there are many lists of interest groups
and lobbyists available {Washington Representatives is probably the most ex-
tensive) , there is no roster of groups that specifically employ the tactic of
stimulating constituent communications to Congress. In fact, the lobbying
reform bill that was finally passed in 1995 specifically excluded requiring
groups to register if they stimulate constituent communications. Further-
more, many groups employ consultants and other third parties to conduct
their grass roots campaigns. So, with no readily available list from which to
sample, I created a sampling frame from different media and political
sources.
First, from April through November 1994, while working in Washington,
I monitored on a daily basis the New York Times, the Wall Street Journal, and
the Washington Post, making note of any ideological group, union, trade
association, or corporation that was mentioned as pursuing grass roots tac-

tics. Second, I also monitored two weekly publications that cover Congress,
interest groups, and professional activity inside the beltway: Congressional
Quarterly Weekly Report and the National Journal. Third, I monitored the
Hotline, a daily briefing memo of political news that summarizes reports
from media outlets from all over the country, and the Healthline, a similar
service that concentrates solely on the policy and politics surrounding health
care legislation.4 Fourth, I obtained a copy of the registration list for the
Public Affairs Council's annual workshop on mobilization tactics. (See Ap-
pendix B for a list of the groups in the sampling frame.)
Using the list derived from these sources, I drew a random sample of
twenty groups from each of the four different types of interest groups and
sent letters requesting confidential, not-for-direct-attribution interviews to
representatives of eighty groups.5 In the case of unions and left-leaning
ideological groups, the initial response to my letters was quite poor. 6 There-
fore, I used the introductions of friends in Washington to gain access to two
of these groups and then used these connections to schedule other inter-
views.7 Even with this snowball method, the interviews were with groups
that were originally on my list. All told, I received responses from forty-eight
groups and eventually interviewed representatives from forty-one for a re-
sponse rate of 51 percent. Table 1.2 shows the distribution of groups in my
sampling frame and final sample, as well as response rates and the number of
grass roots campaigns discussed.8
Since the unit of analysis was individual lobbying campaigns, there was an
added level of sampling. My first question to the interest group representa-
tives I interviewed was to name three recent issues in which they employed
grass roots tactics. Even though my goal was to talk about all the recent
issues where a group pursued a grass roots strategy that attempted to stimu-
4
While the monitoring of all these publications obviously could have been done anywhere,
working in an office that had each of these papers and magazines delivered made the process
much easier.
5
The interview requests were sent on University of Michigan Department of Political Science
stationery, and I identified myself as a graduate student conducting dissertation research.
6
This experience is similar to reports I have heard from other scholars. It is also similar to Jack
Walker's experience with his mail survey of interest groups, where unions had the lowest
response rate and were not used in his analysis (1991).
7
In general, I also found that unions and ideological groups were not as forthcoming in the
interviews. They allowed me to look at fewer background documents and gave me less
detailed behind-the-scenes explanations. In a significant finding for future social science
research methods, the one exception to this rule was the union representative whom I
interviewed at a bar!
8
A recent mass survey of interest groups yielded the following distribution of interest group
types (Leech 1997): trade associations, 28 percent; professional associations, 20 percent;
businesses and corporations, 19 percent; government and institutions, 5 percent; and other
nonprofits, 28 percent.
Table 1.2. Sample Characteristics and Response Rates
Number in Interviews Completed Mobilization

Group Type Sampling Frame Requested Interviews Campaigns
Trade Associations 54 20 10 23
Corporations 55 20 11 16
Citizens' Groups 56 20 12 37
Labor Unions 26 20 8 18
Total 191 80 41 94
Note: In his 1985 survey of interest groups Walker used the Washington Information Directory, published by Congressional
Quarterly Press, for his sampling frame. Because Walker was primarily interested in how organizational factors influence
strategies, he excluded corporations from his sample. For-profit trade associations composed 37.8 percent of his sample,
not-for-profit trade associations composed 32.5 percent, mixed-trade associations composed 5.8 percent, and citizen
groups composed 23.9 percent (Walker 1991, 51).
late constituent communications, time often did not permit this. This sam-
pling method focused on the first campaign that a respondent picked and
likely yielded larger and more prominent campaigns.9
The research strategy of discussing specific issues differs from how some
scholars have previously studied interest group tactics and tactical decisions.
Most previous surveys of interest groups have asked respondents to general-
ize about their activities and the rationale behind their actions (Berry 1977;
Schlozman and Tierney 1986; Walker 1991). Although these surveys pro-
vided extraordinarily valuable information, their methods make it impossi-
ble to understand the political context or political environment in which
lobbying decisions take place. Wanting to understand how the political en-
vironment influences political decisions and believing that respondents
would be best able to describe their actual choices, my method borrows
from Kingdon (1989). Specifically, I asked my informants to talk about their
decision making in the context of particular political issues.10
The great majority of the interviews were conducted in person in the
Washington, D.C. offices of the selected groups. Most of the interviews were
completed between February and November 1995. Three of the interviews,
however, were conducted at corporate headquarters outside the beltway;
three were conducted on the phone; one was conducted in a taxi as I accom-
panied one of my respondents on a trip to Dulles Airport; and one was
conducted at a bar.
The interest group representatives with whom I spoke were familiar with
the rules for not-for-attribution interviews and seemed comfortable with my
note taking. Occasionally - usually after a particularly frank comment -
informants would remind me that their comments were not-for-attribution.
During the interview I coded responses to questions from my interview
schedule about strategic objectives as well as constituent and legislator tar-
geting. In only a few cases did I have to push a respondent to explain to me
more clearly his or her organization's strategic goal or tactical choices
on a particular lobbying campaign. (See Appendix B for the interview
instrument.)
Although my goal was to gather basic quantifiable information about the
maximum number of lobbying choices, I also wanted to give respondents
the opportunity to provide me with more in-depth information and a greater
understanding of how grass roots lobbying campaigns are conducted. In
other words, while still striving to gather basic information from every inter-
9
If I were to carry out a similar study in the future, I would choose a random number before
starting the interview and begin my interview with the "nth" issue mentioned in response to
my first question. This would minimize the focus on the first (and most prominent) cam-
paign that a respondent decided to discuss.
10
A similar strategy was followed by Browne (1988, 1995), and Hansen (1991).
Introduction 11
view, I also wanted to take advantage of the access I had gained and the
knowledge and inside insights of my informants. Therefore, if I was getting
particularly frank, colorful, or more in-depth information, I would not hesi-
tate to deviate from the interview schedule.
To reiterate, the list from which I drew my sample was not an exhaustive
and unbiased one of groups pursuing grass roots tactics. Therefore, the
interview data do not reflect a random sample of all interest groups that
orchestrated constituent communications, nor all tactical decisions that
were made around issues in the 103rd and 104th Congresses.11 Although it
is impossible to measure bias with no information on exactly what the popu-
lation should look like, my sample drawn from media sources surely over-
represents large, prominent, and media-friendly groups. Consequently, it is
impossible to make inferences about all groups, all mobilization campaigns,
or all tactical choices from the frequencies in my data. What's more, the data
do not permit judgments to be made about the effectiveness of grass roots
mobilization as a lobbying tactic.
Nevertheless, whereas in a perfect world a random sample of groups
pursuing grass roots tactics would have been readily available, the goal of
this research was not to measure the effect, frequency, or even organiza-
tional factors that allow grass roots campaigns to be pursued. Instead, the
goal of this research was to study strategic calculations and political decision
making. Although no claim is made that they perfectly represent the entire
universe of groups pursuing grass roots mobilization strategies, I sampled
and completed interviews with a wide range of groups holding diverse ide-
ologies, goals, and organizational structures. All in all, the data from these
interviews provided a substantial first step toward understanding how and
why interest groups stimulate the grass roots.
Taking a page from Richard Fenno (1978), many of the arguments made
in this book are also informed by a significant amount of soaking and pok-
ing. For instance, over an eight-month period during 1994, I worked as a
participant-observer pollster with a political consulting firm in Washington,
D.C. This experience provided me with a unique vantage point from which
to view electoral and legislative politics preceding the 1994 election. The
firm's client list not only included many Democratic congressional candi-
dates, but scores of interest groups on both sides of every imaginable
fence.12 My experience working in Washington allowed me to immerse
11
Although my sampling frame was comprised of groups pursuing grass roots strategies in the
103rd Congress, many of these groups also pursued similar tactics in the 104th Congress.
Therefore, my sample of grass roots campaigns also includes issues from the 104th
Congress.
12
From May to November of 1994 I worked as a senior analyst for the Washington, D.C,
polling firm of Mellman-Lazarus-Lake. Its client list included four incumbent Democrats
myself in the political process - to observe and talk with members of Con-
gress, interest group representatives, and congressional and White House
staff, as well as other consultants. In addition, I drew on information gath-
ered at workshops for grass roots professionals and my own experiences and
observations as a journalist to understand the strategies and tactics as well as
the methods and technologies involved in stimulating constituent com-
munications to Congress.
Plan of the Book

In Chapter 2, I trace the rise of communications to Congress and briefly
discuss some previous work on lobbying and participation. I discuss how the
traditional explanations for participation have a difficult time accounting for
differences in rates of communication to Congress from year to year and
district to district (not to mention why any citizens contact Congress at all).
Although mobilization by elites appears to be a possible solution and is on
the rise as a lobbying tactic, I also discuss how traditional explanations of
lobbying tell us little about when, where, and toward whom is the tactic
likely to be used.
In Chapter 3,1 draw on the participation, legislative behavior, and inter-
est group literatures to devise a theoretical model to understand why and
how strategic and grass roots lobbying decisions are made. I argue that
lobbyists have varied motives and that their tactical choices depend on their
strategic objectives and the information they must convey to legislators and
constituents. I argue that targeting decisions are a multistage process in
which lobbyists attempt to evaluate the influence that a particular individ-
ual's communication to Congress will have on the eventual outcome of a
particular legislative or electoral fight. These tactical judgments ultimately
determine decisions about when to target, where to target, whom to target,
and how to target. In building a theory, I argue that these judgments are
based on the fact that, unlike votes in an election, communications to Con-
gress are not interchangeable. More precisely, I argue that the effect of a
communication on the policy process is a function of the individual com-
who lost their bids for reelection: Speaker Tom Foley (Washington), Rep. Jolene Unsoeld
(Washington), Rep. David Price (North Carolina), and Rep. Larry LaRocco (Idaho), as
well as the House Democratic Caucus. In addition, two Democratic incumbents who won
by only a few thousand votes, Rep. Sander Levin (Michigan) and Rep. Elizabeth Furse
(Oregon), were also clients. While this was unfortunate for the firm, it provided an excellent
vantage point from which to witness the forces that defeated Democratic incumbents in
1994. The firm's client list also included current House Minority Leader Dick Gephardt
(Missouri) and Senate Minority Leader Tom Daschle (South Dakota). Interest group
clients included COPE, Human Rights Campaign Fund, and the League of Conservation
Voters. Corporate clients included chemical giant Freeport McMoran and the five largest
insurance companies.
Introduction 13
municating, the legislator being communicated to, and the message needing
to be communicated.
In Chapter 4, I begin the empirical tests of my theory of grass roots
lobbying choices. I use data gathered from seventy-three grass roots lobby-
ing campaigns to examine when, where, and how groups recruit citizens into
politics, and who gets recruited. In addition to the interviews, I reviewed
primary documents and media accounts and made use of my own experi-
ences as a journalist and political consultant.
Chapter 5 turns to the specific case of health care. The debate over health
care reform and the Clinton plan, which would have changed the way one-
seventh of the economy functions, was one of the biggest policy battles in
recent memory. Using a combination of media accounts, professional pub-
lications, and interviews with many of the major policy and lobbying players,
I examine the role elite-orchestrated mass participation played in the battle
over health care reform. I use this case as a way to test the assumptions from
the theoretical framework as well as a way to acquire information about
legislative and citizen targets during the summer of 1994. Such information
about targeted states and districts is critical in testing a strategic model of
mobilization and how participation between elections works.
In Chapter 6,1 utilize the two sets of survey data - a 1994 Battleground
Poll and a 1994 Times-Mirror poll - to explore patterns of lobbying and
participation. More precisely, using econometric modeling in conjunction
with the targeting information gathered in Chapter 5,1 gauge the indepen-
dent effect of group-targeting strategies on individual behavior. Examining
individuals in cross section provides information about the effect that re-
sources, characteristics, attitudes, and contacts with political leaders have
on an individual's likelihood of participating. Adding contextual variables
on targeted districts allows me to vary the strategic situation in order to
gauge how the political environment influenced both elite recruitment and
citizen participation.
Chapter 7, the concluding chapter, provides me with an opportunity not
only to assess my theory in light of empirical findings but also to place the
nature and use of elite mobilization of mass participation into a broader
context. I discuss how my findings fit into the interest group and mass
participation literatures and explain what my findings mean for specific
reforms that have been proposed.
Patterns and Puzzles in Participation
and Lobbying
The reigning theories of participation in American

politics, amazing as it may seem, do not have much
to say about politics. Instead, they trace activism to
the characteristics of individual American citizens,
to their educations, their incomes, and their
efficacy. They assume that attitudes determine be-
havior. When asked to account for changes in
citizen involvement over the last half century, these
explanations largely fail.
Steven Rosenstone and Mark Hansen, Mobilization,
Participation, and Democracy in America
When he was first elected to the Senate in 1958, the late Philip Hart of
Michigan received fewer than two thousand letters on issues each month. By
1975, near the end of his Senate career, he was receiving approximately ten
thousand letters per month.1 In a typical month in 1995, another Michigan
Senator, Carl Levin, received more than twenty thousand issue letters, along
with five thousand telegrams, ten thousand phone calls, and one thousand
faxes.2
The increase in constituent communications to Congress over the last
three decades has not been confined to Michigan senators. As Figure 2.1
illustrates, there has been a dramatic increase in the amount of mail de-
livered to the House of Representatives over the past thirty years.3
1
Author's calculations from the Philip Hart Papers, Bentley Historical Library, University of
Michigan, Ann Arbor.
2
Author's correspondence with Senator Carl Levin's office. These figures do not include
other types of communications that a senator's office receives - such as requests for service
and information - but just communications that express constituents' views on issues.
3
Calculating or tracking over time the volume of constituent communications that convey
citizen attitudes on issues to members of Congress is not an easy task. Students of elections
have virtually uniform access to precinct, town, county, congressional district, and state
results for the past hundred years. Although there may have been some fraud here or there
14
Patterns and Puzzles 15
Mirroring this pattern, more than nine out of ten respondents (91 per-
cent) to a survey I conducted of congressional offices reported record levels
of incoming communications in 1993 and 1994. All told, almost 160 million
pieces of mail (157,737,234 to be exact) were delivered to the House and
Senate during the 103rd session of Congress according to the Congressional
Research Service (John Pontius, personal correspondence, 1995). In addi-
tion, millions more Americans phoned, faxed, sent telegrams, and e-mailed
Capitol Hill to convey their feelings on issues. My survey of congressional
offices indicated that 43 percent of constituent communications were let-
ters, 41 percent were phone calls, 12 percent were faxes, and 4 percent were
and some observers have questioned the methods by which eligible voters are counted,
debates about the proportion of Americans voting range within 1 or 2 percent. No such
statistics exist for constituent communications to Congress and estimates of what proportion
of American citizens contact Congress to convey their feelings on issues vary greatly. More-
over, with elections, we not only know how many citizens voted, but what they voted on and
whom they voted for. The House and Senate keep track of total incoming mail per month,
but it is impossible to identify to whom it was sent, what it said, and whether it was issue mail.
Mail counts also do not include other ways in which constituents communicate with their
representatives, and thus may understate the level of constituent communications on issues.
Media reports and my own observations in Washington suggest that faxes, phone calls,
telegrams, and even e-mails are increasingly the tools that citizens use to convey their feelings
to Congress. Nevertheless, while these problems may understate participation, we must also
remember that unlike voting, citizens can communicate with Congress as many times as they
want. In sum, aggregate measures of constituent communications to Congress are crude
ones that may tell us little about individual rates of participation.
There are also problems when we attempt to gauge rates of participation from the view-
point of the individual. In the same way that public opinion surveys overstate turnout
because of overreporting and the exclusion of low-probability voters from most sampling
frames, public opinion surveys surely overstate the number of people who contact Congress.
Also, although the "did you vote" question yields socially desirable responses, it is free of
other types of measurement error (it is pretty clear what behavior the question is trying to
gauge and when that behavior was done).
Since no data are available on the volume and character of incoming communications to
individual congressional offices, I conducted a survey of congressional offices. Not requiring
the responses of actual members, the survey was designed to be answered by staff members
familiar with patterns of incoming communications to the office. In addition, to capture
information from members who retired or who lost their bid for reelection in 1994, I
conducted a phone survey of the offices of outgoing members of Congress immediately after
the 1994 election. Unfortunately, the response to the initial round of the mail survey was not
good (52 out 412).
A research assistant and I conducted a follow-up survey by phone of key committee
members in the House and of the entire Senate during the fall of 1995. All in all, I received
information from 112 separate House offices and 47 Senate offices. The data provided some
basic information about increases and the nature of incoming communications. However,
concerns about the reliability and comparability of the responses and the different ways that
offices track their incoming communications make it impossible to use the data for any more
in-depth analysis.
For example, some offices provided me with estimates while others with actual numbers.
Some offices gave percentages, some gave raw numbers. Some counted postcards, while
others did not. Some provided answers for the years asked, but many gave responses for
incorrect years.
Figure 2.1. Mail to House of Representatives, 1973-1994. Source: Mann and

Ornstein 1994, 164; Congressional Research Service (John Pontius, 1995).
telegrams. Also, according to my survey of congressional offices, more than

three-quarters (77 percent) of the communications were from citizens con-
veying their feelings on issues.
Unfortunately, using these data, it is virtually impossible to distinguish
whether this increase has been due to more citizens communicating to Con-
gress or the same citizens communicating their feelings more often. Anecdo-
tal evidence and conversations with longtime Hill staffers suggest that a
significant part of the increase has been greater volume from the same
citizens. Yet, these same staffers also believe that more citizens are writing.
The notion that more individuals are contacting Congress is confirmed by
National Election Study data indicating that there has been an increase in
the proportion of American's contacting their representatives in Washington
to convey their feelings on issues.4 The recent pattern is shown in Figure
2.2.
The NES asks a two-stage question to gauge whether respondents contacted their member of
Congress to convey their feelings on an issue. First, NES asks whether a respondent con-
tacted his or her representative for any reason, and then specifically asks whether a respon-
dent contacted the representative to convey their feelings on an issue. However, NES asks
only whether a respondent contacted his or her representative in the House. The years
included in Figure 2.2 are the only years in which NES asked the question of all respondents
and not only of those residing in a district where an incumbent was running for reelection.
10 n
1978 1980 1986 1990 1994

Figure 2.2 Percentage of citizens contacting Congress to express opinion on
issues, 1978-1994. For coding, see Appendix A. Source: American National
Election Studies.
Using the Participation in America data set as a baseline, Verba, Schloz-

man, and Brady compared rates of participation from 1967 to 1987. They
found that the proportion of Americans reporting an issue-based contact
with a national official doubled, from 11 percent of the population in 1967
to 22 percent of the population in 1987 (Verba et al. 1995a, 72).
Although Roper Center surveys (the longest available time series of com-
munications to Congress) indicate a drop in "letter writing to Congress"
over the past twenty years, the Roper surveys do not measure other ways in
which citizens may contact their members of Congress. For example, Roper
Center surveys conducted during the summer of 1994 found that only a
little over one in ten Americans (12 percent) had written Congress in the last
year.5 However, Times-Mirror Center surveys taken during the identical
time period showed that over one in four Americans (26 percent) reported
having called, faxed, or written a letter to their congressional representative
or senator in the past year.6 Thus, when the survey question is expanded to
include more modern modes of communication, rates of participation indi-
cate a sharp rise.
Although it is virtually impossible to calculate exactly how many more
individuals are communicating to Congress, data suggest a general increase
in both the number of constituents communicating and the volume of con-
stituent communications to Congress. Overlaid upon the trend of increasing
5
Roper Center Surveys June, July, and August (N = 2,000).
6
Times-Mirror Survey, June 23-26 (N = 1,021); Times-Mirror Survey, July 12-25 (N =
3,800).
communications to Congress over the past two decades are variations from
year to year, district to district, and state to state. How can we account for
both the general increase and these the accompanying fluctuations?
In one of the more robust empirical findings in political science, we know
that higher socioeconomic status - specifically, higher levels of education -
help citizens meet the nontrivial costs of participation and spur higher rates
of political activity (Conway 1991; Milbrath and Goel 1977; Verba and Nie
1972; Verba, Schlozman, and Brady 1995a; Wolfinger and Rosenstone
1980). Yet increases in education levels alone cannot explain the rise in
constituent communications to Congress that has apparently taken place
over the past two decades. Moreover, they cannot explain variations from
district to district and state to state.
Partisan attachments (Abramson and Aldrich 1982; Campbell et al.
1960) along with political interest (Neumann 1986; Verba and Nie 1972),
efficacy (Abramson 1983; Bennet 1986), and information (Zaller 1992)
have also been shown to be potent predictors of political activity. Neverthe-
less, at the same time there have been increases in rates of communications
to Congress, all available evidence points to decreases or little change in
partisan attachments, political interest, political efficacy, and levels of politi-
cal information.7
In short, the most prominent schools of thought are unable to account
fully for variation in rates of participation from time to time and place to
place. It's not that this previous scholarship on the correlates of participa-
tion is incorrect. Rather, it is incomplete. Individual demographic and at-
titudinal characteristics are a part of the participation story but not the
entire story.
The Paradox of Participation

Some citizens participate more than others and certain citizens participate
only at certain times. A deeper question that many scholars have attempted
to explain is why there is any mass participation at all? Virtually any type of
mass political participation seems to defy rational explanation. To an indi-
vidual, the costs of participation would appear to outweigh the impact that
one's participation could have on the outcome, so why bother to participate?
That is to say, since individuals can share in "collective" or "public" goods
even if they fail to contribute to their provision, there is little incentive to
engage in most types of mass participation. If my favored legislation passes, I
still share the benefits regardless of whether I wrote my member of Con-
7
Another part of the puzzle is that this increase occurred while rates of voting in congressional
elections have declined.
gress. Only if the desired good is a "private" one, in which my enjoyment of
the benefits hinges upon my participation, can we say that my involvement is
rational.8 If not, I have a strong incentive to be a free rider and a strong
disincentive to bother to communicate with my member of Congress. This
logic is represented by the following familiar formulation (Aldrich 1993;
Downs 1957; Riker and Ordeshook 1968):
R = PB - C
where B stands for the benefits one will receive if one's desired outcome
results, C represents the costs if one chooses to participate, and P stands for
the probability that one's participation will be decisive in achieving the
desired outcome. An individual participates if R is positive. Since P is usu-
ally very small and costs are not trivial, it is difficult to explain why people
participate at all.
Previous findings about the correlates of participation do not provide a
solution to the paradox of participation. In fact, they add to the puzzle. As
Verba, Schlozman, and Brady argue:
Indeed, with respect to SES [socioeconomic status] and participation, at least one
variant of the rational choice approach suggests no relationship at all or, if any, that
people of high SES (who by virtue of their high levels of education command the
intellectual sophistication to comprehend the free rider problem and by virtue of
their high salaries would find the opportunity cost of participation prohibitive) to be
the least likely to take part in politics. Instead those with high levels of SES, who are
not otherwise known for particular irrationality in the conduct of their lives, are likely
the most active. (1995b, 272)
Yet, just as a bumblebee flies despite the "fact" that aeronautical engi-
neers tell us it cannot, people do participate in politics. To get around this
inconvenient empirical finding, rational choice theorists have been forced to
rely on other conceptions of benefits, such as the satisfaction gained from
doing one's civic duty. However, this addition to the theory leads to a tautol-
ogy in which people who participate in politics must receive civic satisfaction
for doing so, while those who do not participate get no civic satisfaction
from participating.
Nevertheless, these flaws do not destroy the utility of the rational choice
calculus of participation. In fact, it has proved to be a useful tool in structur-
ing arguments about participation and in identifying which factors influence
individual participation. For instance, in the electoral context, more com-
petitive races (with a greater chance of one's vote being decisive) have been
demonstrated to increase voter turnout (Goldstein 1994; Rosenstone and
Hansen 1993). This finding does not imply that individuals calculate their
8
See Olson (1965) for the classic discussion of the collective action problem.
individual probabilities of influencing an election race. In fact, studies indi-

cate people are no more likely to cast their ballots when they perceive the
race to be a close one (Gant and Luttbeg 1991). Why then is turnout higher
in close elections?
The answer appears to be that political leaders and campaign workers are
encouraged by close races to work harder and to stimulate more voters.
Thus closeness exerts an indirect, yet powerful, effect on an individual's
chance of turning out to vote. Individual citizens may not care if an election
is close; but parties, groups, and campaigns seek to invest resources where
they will do the most good. Therefore, they are more likely to mobilize
participation when a race is close. In this manner, although rational choice
theories may not do a particularly good job in directly predicting individual
participation choices, they may do a very good job in predicting campaign
choices and in structuring our understanding of how the strategic environ-
ment can influence voter participation.
In Chapter 3,1 argue that a similar logic helps us explain how organized
interests make tactical decisions about lobbying between elections - about
which legislators need to be targeted, which citizens need to be recruited,
and what messages need to be conveyed. In a lobbying context, however, the
costs, benefits, and probabilities represented in the rational calculus of a
participatory decision are not identical for all individuals, all types of par-
ticipation, or all types of political environments. The nature of probabilities,
benefits, and costs will vary for different individuals who are deciding
whether or not to contact Congress. These calculations are crucial as politi-
cal leaders set out to make tactical decisions. Even if people don't use
rational calculations in deciding whether or not to participate in politics,
political leaders design their strategies and invest their resources as if they
do.
A Mobilization Solution
In attempting to untangle the correlates of citizen participation, Verba,
Schlozman, and Brady (1995a) distinguished between long-term causes
(individual attitudes, attributes, and characteristics) and short-term causes,
such as attitudes on issues. Not surprisingly, since they conducted their
survey in the wake of the Supreme Court's Wkbster decision and the reentry of
the abortion debate into the legislative battleground, the authors found that
attitudes on abortion helped drive communications to Congress. The au-
thors, however, generally discounted the influence of short-term factors in
favor of a model that relies on more long-term explanations of participation.
In contrast to the more or less standing decision embodied by the Civic Voluntarism
Model, issue engagements constitute a wild card with respect to their impact on
participatory stratification. Their consequences for the representation of publics

otherwise not well represented through participation are not fixed and stable. . . . As
issues come and go, they mobilize to politics different issue publics. Because the
issues that engender issue engagements are many and transitory, it is impossible to
assess their net effect in reinforcing or counteracting the unambiguous impact of the
structural factors in the Civic Voluntarism Model in creating participatory distor-
tion. Nevertheless, the additional boost given to activity by an issue engagement is a
small increment to a level of activity shaped more fundamentally by long-term fac-
tors. (Verba, Schlozman, and Brady 1995a, 522)
Earlier in their study, Verba, Schlozman, and Brady do demonstrate the

effect that requests for political activity have on stimulating communications
on policy issues. This finding is consistent with scores of previous studies on
the impact of elite mobilization on electoral turnout (Cutright 1963; Gos-
nell 1927; Katz and Eldersveld 1961; Kramer 1970; Nagel 1987; Powell
1986; Wolfinger 1963). Still they argue, "The process by which people
become the target of requests for activity is socially structured" (Verba et al.
1996, 149).
Although I agree that requests for participation are socially structured, I
also argue that the process is politically structured. Short of understanding
the "transitory" ways in which issues come on the agenda, short-term
factors - political factors - can still help us understand political participa-
tion. Long-term factors certainly have a large influence on shaping par-
ticipation and, as I argue, on tactical lobbying choices. The effect, however,
of short-term factors is not simply a "wild card" having trivial, unsystematic
effects. Political participation is a political act, with political goals, causes,
and consequences. Issues alone do not mobilize citizens to political activity.
Political leaders recruit citizens to political activity for political reasons.
Rosenstone and Hansen (1993) made a similar argument. They claimed
that individual attributes cannot change quickly enough to explain changes
in rates of political participation and that changes in elite patterns of mobili-
zation can explain much of the variance in rates of participation. Although
much attention has been paid to their claims about the role of mobilization
in influencing voting and explaining the historical decline in turnout, Rosen-
stone and Hansen have also made a much broader argument about the effect
of mobilization on political participation in general.
Specifically, they theorized that activities such as letter writing, petition
signing, and demonstrating, dubbed "governmental participation," fluctu-
ated according to the strategic decisions of politicians, interest groups, and
parties. Using a thirty-year time series of Roper Survey data in addition to a
cross section from the 1976 National Election Study, Rosenstone and
Hansen demonstrated that the strategic choices of politicians, political par-
ties, interest groups, and activists play a crucial role in determining who
participates in politics and when they participate in politics. They argue that
"Personal characteristics - resources, perceived rewards, interests, and ben-

efits from taking part in politics - define every person's predisposition to-
ward political activity. The strategic choices of political leaders - their deter-
mination of who and when to mobilize - determine the shape of political
participation in America" (Rosenstone and Hansen 1993, 36).9
If requests for political activity have such a powerful effect on individual
behavior, we must understand what drives these decisions. We need to
understand the less proximate causes of participation, the tactical decisions
of political leaders. In other words, if requests by political leaders help drive
participation, we must ask and answer the antecedent question: what drives
these requests?
Interest Group Lobbying Tactics

Alongside the increase in constituent communications to Congress and
consistent with the preceding discussion on the importance of recruitment
and requests for participation has been a contemporaneous increase in the
use of grass roots lobbying tactics. Specifically, interest groups and lobbying
firms inside the beltway are increasingly making use of new and more so-
phisticated technologies to water the grass roots outside the beltway. To be
sure, stimulating the grass roots is not a new technique. For example,
around the turn of the century, the Anti-Saloon League had a mailing list of
over 500,000 people and was the moving force behind the passage of pro-
hibition (Odegard 1928). Writing in the 1920s, Pendleton Herring (1929)
described the grass roots activities of corporations and interest groups dur-
ing that time period. Going back even farther, Margaret Thomson (1985)
9
Rosenstone and Hansen develop a convincing theoretical and empirical argument about the
role of elite mobilization in mass participation. Unfortunately, in their models of "govern-
mental participation," they face problems in operationalizing and measuring mobilization.
In their examination of mobilization and voting, they are able to utilize specific questions to
gauge whether a respondent was mobilized and are able to augment these data with contex-
tual variables on competitive states and districts (where rational elites should concentrate
their resources and mobilization efforts). However, in modeling the NES cross section and
the Roper Center time series on nonelectoral participation and mobilization, they have
neither direct measures of mobilization at the individual level nor direct measures of where
mobilization efforts are likely to be targeted. The following variables were used as measures
of mobilization in governmental politics: percentage of population that worked for political
party; percentage of senators and representatives serving first two years of term; the natural
log of number of bills introduced in year; presidential speeches in past week requesting letters
to Congress; unemployment rate; presidential budget request as a percentage of last year's
budget; number of days in session in past four weeks; and the natural log of the number of
bills on the floor (Rosenstone and Hansen 1993, 266). These are creative measures designed
to address a difficult problem and are likely measuring mobilization to some extent. How-
ever, all they measure with these variables can not be solely attributed to mobilization. For
example, number of bills could be mobilization or it could just be that there are more
opportunities or issues on which to write letters. Just as an election is an opportunity to
mobilize, it is also an opportunity to vote.
found that corporations in the 1870s used grass roots lobbying tactics to
curry favor with federal officials.
In their landmark study of lobbying and congressional decision making,
Bauer, Pool, and Dexter (1964) described instances where particular indus-
tries and groups attempted to recruit employees and association members to
write in support of, or opposition to, provisions of the Reciprocal Trade Act.
Although generally skeptical about the power of interest groups to influence
public policy, the authors described instances in which industry-orche-
strated grass roots efforts apparently influenced legislative outcomes. In
particular, they described how the American Cotton Manufacturers In-
stitute (ACMI) organized millowners and workers to undertake a massive
pro-protectionist letter-writing campaign directed at formally free-trade
southern congressmen. The ACMI did not directly lobby southern con-
gressmen, but attempted to influence them indirectly by recruiting influen-
tial constituents. The result was a shift in positions among traditionally free-
trading southern congressman. Although recruiting influential constituents
was apparently effective, Bauer, Pool, and Dexter opined, "the methods
used by the ACMI to produce the 1955 shift were classic and unoriginal"
(1964, 354).
More recently, Schlozman and Tierney argued that the increase in indi-
rect lobbying techniques that occurred in the early 1980s was simply part of
a broader increase in all sorts of lobbying activities. Respondents to Schloz-
man and Tierney's survey of Washington organizations reported using a
wide variety of techniques and using them more frequently. In short, Schloz-
man and Tierney described a world in which the grass roots strategies and
other lobbying techniques utilized by groups were simply "more of the
same" (1983, 1986).
Still, of all the types of lobbying tactics examined in their study, the
greatest increases occurred in outside tactics: "talking with people from the
press and the media," "mounting grass roots campaigns," and "inspiring
letter-writing" or "telegram campaigns" (Schlozman and Tierney 1983,
1986). Also, the fact that a tactic may be "unoriginal" or just "more of the
same" does not mean it is ineffective or well understood by scholars. Fur-
thermore, as Cigler and Loomis argue, "More of the same becomes at some
point something categorically different" (1995, 393). Tierney himself ad-
mits in a later work, "In politics, as in oil spills or hazardous wastes, more of
the same is not really the same" (1992, 219).
If the 1980s were more of the same when it came to grass roots lobbying,
the early 1990s have been much more of the same. That is, it appears that
grass roots lobbying is being used in different, more sophisticated, and
potentially more powerful ways.10 What can explain this increase?
10
In Chapter 4, I discuss technological advances in much greater detail.
Kernell (1993) explained the rise in "going public" by presidents as a

function of Washington politics moving from a system of institutional plural-
ism, in which bargains were struck by a few leaders of protocoalitions, to one
of individual pluralism, where decisions were contested by an increasing
number of free agents. The decrease in bargaining and the proliferation of
outside strategies can also explain the rise in interest groups going public.
The proliferation of subcommittees, the loss of power by committee chairs,
and the weakening of the national parties demands that those who seek to
influence public policy must now influence more policy makers. These same
factors have also led to increasingly independent legislators who rely on their
own actions, accomplishments, and skills for political survival. Further-
more, the long reach of the federal government and the rise of the modern
welfare state have created constituencies outside the beltway that have an
interest in the policies made inside the beltway. In addition, the growth of
the mass media has made it easier for representatives and constituents to
communicate with each other, and technological advances have made it
feasible for groups to generate, virtually instantaneously, thousands and
even hundreds of thousands of letters, faxes, phone calls, and telegrams
when an issue or bill comes to a head.
Although it is difficult to measure precisely how much more frequently
grass roots lobbying tactics are being used, and even more difficult to gauge
their growth against some sort of baseline, there is at least some evidence,
aside from anecdotal reports, to indicate that there has been a change in the
magnitude and nature of such indirect tactics. For instance, media coverage
of grass roots mobilization as a lobbying tactic has increased. Using on-line
databases, I conducted a story count of the number of articles in the Na-
tional Journal, the New York Times, and the Washington Post that contained
the phrase "grass roots lobbying."11 As Figure 2.3 illustrates, the number of
articles about grass roots lobbying has increased dramatically during the
past four years. After slowly rising throughout the 1980s, the number of
articles mentioning grass roots lobbying tripled in the early 1990s.
Growth in the use of grass roots tactics by corporations and trade associa-
tions is apparently one major factor driving the increase. Whereas in the
past American business and other interest groups were content to surrender
the grass roots to environmentalists and other citizen groups, they have
clearly learned about the power constituent pressure can wield on policy
makers and have worked hard to influence policy through the use of sophis-
ticated grass roots strategies (Dowd 1993; Mintz 1994; Vogel 1989). The
11
The National Journal is the trade publication that most comprehensively covers lobbying
and the behavior of organized interests in Washington, D.C. This finding should be treated
with caution since different terms may have been used to describe political recruitment over
the past twenty years.
New York Times

160 1
Washington Post
140 Natio nal Jo ur nal
120 "
100 -
80
60 1
40
20
0
77 79 81 83 85 87 89 91 93
Figure 2. 3. Grass roots lobbying mentions in the press, 1977-1994. Source:

Nexis.
Public Affairs Council, a Washington-based group of corporate and trade

association professionals, reports dramatic increases in the number of at-
tendees at its annual conference on grass roots government relations tech-
niques (Tony Kramer, personal interview, 1995).
Also, grass roots lobbying tactics are the subject about which the Public
Affairs Council receives the most inquires. Moreover, most Fortune 500
companies now have full-time grass roots organizers and detailed plans of
action that they can implement when needed. Efforts that were once ad hoc
have now become permanent and official divisions of many corporations
(Swift-Rosensweig 1995).
In addition, a whole new cadre of consultants who specialize in or-
chestrating mass participation has cropped up. Professional directories from
Campaigns and Elections magazine show that the number of grass roots firms
has grown every year and that more than 150 firms now specialize in stirring
up mass participation.
These trends have encouraged a countermobilization (or a counter-
countermobilization) by environmental and labor groups. Following the
1994 elections, the American Federation of State, County and Municipal
Employees spearheaded a drive by organized labor - Project 95 - to place
field organizers in thirty swing congressional districts. Similarly, the League
of Conservation Voters placed field organizers in twenty swing congressional
districts. During the August recess of 1995, the Democratic National Com-
mittee ran television advertisements about Medicare in thirteen targeted
congressional districts that encouraged viewers to contact their member of
Congress. At the same time the AFL-CIO conducted a million-dollar televi-
sion and radio campaign aimed at twenty-six House districts {Hotline, Au-
gust 23, 1995).12
Issue Advocacy and Independent Expenditure Campaigns

There is an important distinction between lobbying and advertising cam-
paigns conducted more than a year before the election and the activities of
groups during the electoral season. Think about the following campaigns
conducted during the summer and fall of 1996.
Sierra Club, Citizens Action, and the League of Conservation Voters
aired the following commercial during the summer of 1996 in the congres-
sional districts of fifteen freshmen House Republicans. Over a visual of
beautiful rivers and lakes changing to a visual of polluted lakes and rivers,
the narrator made the following plea:
It's our land; our water. America's environment must be protected. But in just
eighteen months, Congressman [name] has voted twelve out of twelve times to
weaken environmental protections. Congressman [name] even voted to let oil cor-
porations continue releasing cancer-causing pollutants into our air. He voted for the
corporations who lobbied these bills and gave him thousands of dollars. Call Con-
gressman (name) and tell him to protect America's environment. For our families,
for our future. {Hotline, July 12, 1996)
There was no particular piece of legislation in front of Congress when
these advertisements aired, and the call for action did not speak of any
current legislative battle. Phone calls from environmentalists were not going
to change the minds of these representatives, and there was nothing for them
to change their minds about. The request for calls to Congress qualified the
advertisements as issue advertising and provided a way for tax-exempt [501]
(c) (4) organizations to engage in election advertising.
The respective parties' national committee also got in on the act. One of
the most widely run was an advertisement by the Democratic National
Committee. It featured ominous music and fuzzy black and white pictures
of Newt Gingrich and Bob Dole vowing to let Medicare "wither on the
vine." At the end of the commercial a phone number was quickly flashed
and viewers were asked to get in touch with Congress. But the advertisement
and the request for phone calls were not done to influence Dole or Gingrich.
They were done so that these expenditures would qualify as legislative ad-
12
John Sweeney ran and was elected to the presidency of the AFL-CIO largely on a platform
of making organized labor a grass roots force.
vertising. The advertisements were targeted to swing states in the presiden-
tial contest and not swing districts for a legislative debate. They were vehi-
cles for parties to communicate election messages before the campaign
began and to evade campaign spending limits.
The GOP followed a similar strategy. The Republican National Commit-
tee aired an advertisement about Clinton "flip-flopping" on the number
of years it would take to reach a balanced budget. The commercial ended
with a quick request to get in touch with "your elected official." The RNC
also aired the following commercial targeted at battleground states in the
Midwest.
[Woman and child are shown in doctor's office, with child on a nebulizer]
Woman: I can't change jobs without losing my health insurance. Robbie has asthma.
[U.S. Capitol dissolves to shots of workers and patients being treated by doctor]
Announcer: The Republican Congress wants to make sure workers who lose or
change jobs keep their insurance - regardless of their medical condition. But
liberal Democrats are blocking this commonsense approach [picture of Ted Kennedy
and Bill Clinton]. And instead of giving people more choices, they still want govern-
ment-run health care.
Woman: When are they going to learn that's not what we need.
Announcer: Write President Clinton and Senator Ted Kennedy today. Tell them to
support the Republicans' commonsense health care reform.
It was unlikely that Ted Kennedy or Bill Clinton got much useful informa-
tion from any communications stimulated by this advertisement. Once
again, the goal was not to provide information to legislators or President
Clinton, but to provide information to voters and frame particular issues.
The goal was to put Clinton's picture next to Kennedy's in battleground
states without exceeding federal spending limits.
My favorite example of a campaign advertisement dressed up as an issue-
based grass roots lobbying campaign comes from the 1996 House race in the
state of Montana. Citizens for Reform, a conservative group, aired the
following commercial in the middle of the fall campaign (Carney 1997,
1643). The target of this commercial, Democratic candidate Bill Yellowtail,
was not even an incumbent member of the House of Representatives. "He
preaches family values, but he takes a swing at his wife. Yellowtail's explana-
tion? He only slapped her. Call Bill Yellowtail and tell him you don't approve
of his wrongful behavior."
Since this advertisement did not explicitly advocate the election of a par-
ticular candidate, it counted as an issue advocacy campaign, an activity in
which tax-exempt groups are permitted to engage. Peter Flaherty, head of
Citizens for Reform, explained, "As long as we don't use any express-
advocacy words, anything we do is permissible" (Carney 1997). Similar
reasoning allowed scores of other groups to air millions of dollars of ad-
28 Interest Groups, Lobbying., and Participation
vertisements - ostensibly asking for citizen action on a legislative issue -

during the 1996 campaign season.
The stimulus for this project was a better understanding of how tactical
outside lobbying decisions are made and how these decisions influence mass
participation. Still, I do not examine all types of campaigns that request
mass participation. Although some may consider the preceding advertise-
ments aired in the midst of the 1996 campaign to be part of an issue ad-
vocacy campaigns, I do not. The use of independent expenditure and issue
advocacy campaigns during an election is a topic that deserves further study.
Nevertheless, even though the grass roots lobbying campaigns that I discuss
rely on the electoral connection and may even have an electoral objective,
the logic that governs their use is far different from the logic that governs the
use of independent expenditure campaigns. These types of activities are
clearly just another facet of a short-term electoral campaign and should be
studied as such.
Chapter Summary
Many of the central insights of the mass participation and interest group
literatures are correct, but they are incomplete. On the one hand, under-
standing why individuals participate in politics requires more than an under-
standing of individual attributes and attitudes. On the other hand, assessing
the nature of interest group influence requires one to look beyond financial
donations and the direct activities of Washington-based lobbyists. These
factors alone cannot explain the explosion in citizen communications to
Congress over the past decade. Likewise, traditional explanations of par-
ticipation cannot account for this trend or the differences we see in who
participates from year to year.
Requests for political activity play a crucial role in inspiring communica-
tions to legislators and are a commonly used lobbying tactic wielded by
organized interests. Therefore to understand more fully both the nature of
mass participation and interest group influence, we must have a better
understanding of what drives these requests for political activity.
Nevertheless, based on the participation literature, we really know little
about how tactical mobilization decisions are made. By the same token,
traditional explanations of lobbying and interest group influence have con-
centrated on organizational factors and are unable to explain how grass
roots lobbying is employed. In sum, to understand more fully both why
people contact Congress and how lobbying influences the legislative pro-
cess, we must understand how and why strategic and tactical lobbying
choices are made. Up to this point, the lack of a specific theory about why
and how elites make strategic and tactical lobbying decisions has hampered
our understanding of both mass participation and interest group influence.
In the next chapter^ I take a first step toward remedying that deficiency.
Although I will not tackle directly the question of interest group effective-
ness3 having a theoretical understanding of the logic behind interest group
strategies and tactics is a necessary first step for evaluating their worth.
The Political Logic of Political Decisions
All animals are equal, but some animals are more

equal than others.
George Orwell, Animal Farm
Remember, there are only three people in Congress

who really care what your employees think: their
two senators and their representative.
Mobilization manual from the chemical industry
Most scholarly accounts of interest groups and lobbying have focused on

how organizational factors and resources influence lobbying decisions and
choices of tactics. I start from the premise that political factors also have a
crucial influence. I assume that the architecture of the American political
system also governs interest group actions. Therefore, to dissect the specific
strategies and tactics lobbyists adopt in dealing with Congress, we must first
identify what factors actually influence congressional elections and congres-
sional decisions.
In the first part of this chapter, I take advantage of what scholars have
learned about elections and legislative decision making to understand how
and why grass roots strategies are used in lobbying campaigns. I argue that
grass roots lobbying is an effective tool for interest groups to provide - in the
way most likely to bias the process in their favor - the information required
by voters when they decide for whom to cast a ballot and for legislators when
they decide how to proceed on a particular piece of legislation. In the second
section, I argue that given the choice of stimulating constituent communica-
tions as a tactic, targeting decisions are a function of strategic objectives and
are determined by estimations of an individual's probability of influencing
the process. Given a particular strategic objective or piece of information
that needs to be conveyed, I explain the subsequent choices made by lobby-
ists about when to lobby, where to lobby, whom to lobby, and how to lobby.
30
Political Logic of Political Decisions 31
Finally, in the chapter's concluding section, I derive a set of testable proposi-
tions from my representation of lobbying strategies and tactics.
Influencing Congressional Elections

Whether the objective is to influence current legislative battles or future
elections, the lobbying tactic of stimulating constituent communications
relies on the electoral connection and the manner in which issues affect
electoral outcomes. Consequently, understanding the way incumbents' is-
sue stands can affect congressional elections is an important first step in
developing a theory of how grass roots lobbying decisions are made. Figure
3.1 illustrates the basic factors that have been shown to influence congres-
sional elections.1
Congressional candidates whose party and president have high approval
ratings are more likely to be successful. Democratic candidates in Democra-
tic districts are more likely to be successful. High-quality challengers with
sufficient resources are more likely to be successful. Incumbents who are
free of personal scandals and have voted according to the wishes of their
constituents are more likely to be successful.
Although national factors and the partisanship of the district certainly
have a significant impact on congressional elections, given what we know
about party identification and national events, they also are the factors that
legislators and lobbyists can influence the least. One legislator can do little
to make his or her constituents more Republican or the national economy
stronger. Likewise, one interest group can do little to make constituents
more Democratic or insure the successful completion of a war. Granting
these basic determinants of congressional elections and the inability of inter-
est groups to influence some of the most important determinants, how do
lobbyists go about pursuing an election strategy?
Although much can be explained by these fundamental factors, cam-
paigns are won and lost at the margins. Accordingly, interest groups can and
do contribute money to candidates to conduct their campaigns. Further-
more, as discussed in the previous chapter, groups are increasingly conduct-
ing their own campaigns on behalf of friendly candidates. These campaigns
can run into the millions of dollars and include all the trappings of the
typical modern electoral campaign - television advertisements, mailings,
professional field staff, and phone banks.
There is little subtlety to these supposedly independent electoral efforts or
about contributing money directly to candidates. With these actions and
1
The logic outlined in this figure is obviously grossly simplified and does not attempt to
portray the relative magnitudes of the different factors nor the relationships between them.
In devising it, I relied mainly on the work of Arnold (1990) and Jacobson (1992).
National Factors
• Economy
• Crisis
• Presidential Approval
• Party Approval
Partisanship of District
Challenger Quality Election Outcome
Challenger Resources
Incumbent Behavior
Figure 3.1. Major factors influencing congressional elections.
choice of tactics, interest groups are trying to influence voting behavior in

the short term. Although the growth in independent expenditure campaigns
suggests that this is a fertile area for future research, my goal is not to analyze
how interest groups directly influence election outcomes in the short run. In
this chapter and in this book, I concentrate on the indirect ways in which
groups try to influence elections.2
Incumbent Behavior and Establishing Traceability

There is significant debate about the extent to which issues, particularly
incumbent actions on roll call votes, influence congressional elections. Still,
there are ways for organized interests to create the conditions necessary for
issues and incumbent actions to have an impact. Arnold (1990) argued that
for an issue to become salient enough to influence an election - for the
behavior of the incumbent to be an important factor - three conditions must
2
See Kingdon (1989, chap. 9) for a discussion of how interest groups try to tap into existing
information flows in order to influence congressional decision making. Also see Rothenberg
(1992), Smith (1984), and Wright (1996).
be met to establish what he dubs "traceability." First, there must be percep-
tible effects: gas is more expensive, taxes are higher, or sugar is cheaper.
Second, there must be an identifiable government action that can be linked
plausibly to the particular effect: import quotas on oil are approved, a new
tax bill is passed, or foreign sugar imports are allowed. Third, there must be
a visible contribution by a legislator to that government action: he or she
voted for import quotas, new taxes, or a sugar bill. Arnold (1990) argues
that when these conditions are in place, challengers can capitalize on the
actions of incumbents and particular issues can influence congressional
elections.
My model of how lobbyists influence election outcomes relies very much
on the logic outlined by Arnold (1990). But it differs in one crucial respect.
In Arnold's model, legislative entrepreneurs and the political environment
determine salience and traceability. For example, before an election cam-
paign challengers or instigators determine which issues are salient and on
which issues an incumbent is vulnerable. I argue that more than legislative
entrepreneurs and legislative rules can establish or hide traceability. In my
model, in a legislative fight interest groups seek to define an issue and make
it salient for later use in an electoral fight. Interest groups, however, not only
react in the campaign season when an issue is salient and when traceability
has been established. They also may attempt to establish traceability long
before the formal campaign begins. Along with environmental factors and
the activities of legislative entrepreneurs, interest groups can influence
whether there is traceability on an issue. In other words, interest groups try
to provide the information required for there to be traceability on an issue
and for it to be used as ammunition in a future campaign.
If interest groups do not treat the traceability and salience of an issue as
exogenous, how do lobbyists go about highlighting policy effects and linking
policy outcomes to government actions and incumbent behavior? How do
they go about providing the information necessary to establish traceability?
Although groups can send signals with their monetary contributions and
professional lobbying contacts, stimulating constituent communications to
Congress is another effective way to provide information on electoral conse-
quences. Lobbyists realize that how issues are framed in a legislative fight
influences how they can be framed in a future electoral fight. The legislative
battle over how a policy is defined can reveal much about who eventually
wins the electoral war. For instance, the outcome of the battle between the
parties and their interest group allies about whether a vote for the Republi-
can budget in 1995 was a vote for a balanced budget and lower taxes or a
vote for slashing Medicare to provide tax cuts for the wealthy seems to have
had a significant impact on the 1996 elections. In that battle, mobilizing
hundreds of thousands of constituent communications and conducting a
television campaign in targeted states and districts seems to have been an

effective tool for Democrats to frame the 1996 election during the fall of
1995.
Speaking of other legislative battles that preceded the 1994 election,
David Dixon, a prominent Democratic media consultant and former politi-
cal director of the Democratic Congressional Campaign Committee ex-
plained to me that "It is not enough to research a couple negatives on an
incumbent, test which ones work the best and then make a great ad that airs
a few weeks before the election. You need to be in the field, on the ground,
stirring up opposition, enlisting allies, defining issues and creating defining
moments long before the formal campaign period. Whoever wins the battle
to frame an issue during the legislative battle will be able to use the issue
most effectively in November."
Stimulating communications from the grass roots can be just the way to
stir up opposition, enlist allies, and create defining moments before the
campaign begins. Grass roots lobbying can be a most effective way for
organized interests to frame issues in ways that make policy effects more
salient, that make constituents aware of government actions, and that link
legislators to those policy effects and government actions.
In addition, during a grass roots campaign, messages can be tested and
groups can lay the logistical groundwork for an election campaign. Similar
skills and resources are needed to mobilize people to contact Congress or to
vote. Grass roots campaigns centered on a legislative issue can help groups
to build an infrastructure, train operatives, and gather lists of supporters for
future use in an upcoming election campaign.
Quality Challengers
Congressional elections are also determined by the quality of challengers.
From their intensive case study of a single congressional district in upstate
New York, Fowler and McClure (1989) argued that "The quiet calculations
made by unseen candidates in February or March of a Congressional year or
even earlier have as much to do with the electoral outcome in November as
do the noisy fall campaigns." Simply put, those politicians who are well-
funded and have prior political experience fare better than those who are
poorly funded and have no previous political experience. To a great extent,
who runs determines who wins. So, by recruiting strong candidates sympa-
thetic to their cause and by influencing their "quiet calculations," interest
groups can have an indirect, but powerful, influence on election outcomes.
Yet strong candidates who have the greatest chance of being successful
also face the greatest risks (Jacobson and Kernell 1981). High-quality chal-
Table 3.1. Strategies for Influencing Congressional Elections
Objective Strategy
Influence Congressional Election Define issue and/or establish
traceability for future electoral use
Encourage or discourage potential
challengers
lengers need to be convinced that a race is winnable. Krasno and Green

(1988) demonstrated that local factors, specifically the behavior and per-
sonal vulnerability of incumbents, do influence strategic decisions about
whether to run for Congress.3 Encouraging constituents to contact Con-
gress is one way for interest groups to demonstrate to challengers that an
incumbent is vulnerable. Mobilizing constituents to contact an incumbent
to express their displeasure about a particular action can provide significant
information to challengers. Stimulated constituent communications to
Congress are strong evidence that a winning message exists and that allies
have the resources to deliver it. Stimulated communications may also show
an incumbent that he or she is vulnerable and lead to the decision to retire.
Similarly, stimulating constituents to convey their support for a particular
incumbent's action can encourage an incumbent to run for reelection and
thereby consequently serve to discourage potential challengers.
Summary
Previous scholars had difficulty documenting the direct effect of interest
group activity on congressional elections. I have outlined a model in which
grass roots lobbying can have an indirect, yet powerful, effect (see Table
3.1). Stimulating constituent communications is one way interest groups
can prime the salience of issues and make it possible for an incumbent's
record to be used in the next campaign. In addition, stimulating com-
munications to Congress not only demonstrates to potential challengers that
an incumbent is vulnerable to attack on an issue but demonstrates that
organized groups are able to deliver important blocks of voters.
3
I am agnostic on the debate in the literature about whether it is national or local factors that
most affect strategic calculations about whether a race is winnable. My position is that local
factors do have some influence on a challenger's decision.
Influencing Legislative Action

A logical place to begin to understand how lobbyists attempt to influence
legislative decision making is to understand what, in fact, influences legisla-
tive decision making. As John Kingdon put it, "It is probably a general
strategic principle that those who want something from decision-makers
must adapt their strategies to the decision rules that are being used" (1989,
153).
Legislators have multiple motives when deciding how to act on an issue.
These include the desire to win reelection, the desire to enact good public
policies, and the desire to gain stature within the institution (Fenno 1973).
Thus, when making decisions, constituents, their own policy beliefs, and the
positions of colleagues and party leaders influence legislators. Without
doing too much of a disservice to the volumes of research on congressional
decision making, Figure 3.2 outlines what scholars have taught us about the
forces that influence legislative decision making and the decision rules uti-
lized by legislators. It shows what types of information legislators require
when deciding how to behave on a particular issue. In devising this chart, I
rely mainly on the work of Arnold (1990), Fenno (1978), Kingdon (1989),
and Mayhew (1974).
Consistent with their desire to enact good public policy and gain stature
within the institution, legislators' initial predispositions on a particular issue
or piece of legislation are a function of their personal policy beliefs and the
beliefs of congressional leaders and influential colleagues. Legislators also
want to get reelected, however, and must therefore pay attention to the
opinions of constituents. If a legislator's beliefs are in line with his or her
constituency, then the decision on what action to take is an easy one. Also,
as Figure 3.2 illustrates, when a legislator has no position or the constitu-
ency has no position (or has a divided position), then the decisions are also
fairly straightforward.
Trouble arises when a legislator's initial predispositions are in conflict
with his or her constituents.4 But simply being in conflict is not enough to
influence a legislator's behavior. Claiming that constituency influence mat-
ters and that legislators try to estimate their constituents' policy preferences
does not imply that legislators conduct surveys and vote in the direction of
the top-line results. Even if a member of Congress was capable of polling on
4
Again, no claim is made that incumbent behavior is the most important factor in congressio-
nal elections. In fact, no claim is made about the exact influence of incumbent behavior on
congressional elections. Instead, the more modest claim is that congressional elections are
often won and lost at the margin and that personal actions are one of the few determinants
legislators can control. Therefore, I assume that risk-averse legislators will attempt to maxi-
mize the benefit or minimize the loss from their own actions.
Votes Legislator and Votes Constituency Votes Legislator
Constituency Position Position Position
Constituency Has
No Position or Is
Divided
Legislator Position
Constituency Position
What Is Constituent Intensity?
What Is Legislator Intensity?
High Low
High Redefine Conflict and/or Prefer Legislator's
Consider Other Actors Position
Low Prefer Constituency Position Consider Other Actors
Figure 3.2. Legislative decision making.

every issue, the concern is not with the gross distribution of public opinion
in his or her constituency. Legislators are unlikely to pay attention to even
lopsided margins on an issue poll if they have no evidence that an action will
cause politically important supporters to work against them.
For legislators to be convinced that a particular action could cause future
electoral troubles, they need to be convinced not only about the general level
of public opinion in their district but also that attitudes on a particular issue
could potentially become salient in an electoral campaign and ultimately be
translated into votes. Kingdon used the table in the lower right-hand corner
of Figure 3.2 to illustrate how constituency influence and the electoral
connection structure the choices of legislators when they disagree with con-
stituents on an issue (1989, 39). If a legislator is not convinced that constitu-
ent opinion is intense (that opinions on an issue could be translated into
votes), he or she is free to act his or her own way.
The level of pressure by party leaders and the depth of their own beliefs
determine the intensity of the legislators' position, but how do legislators
judge constituent intensity? How do legislators determine whether their
action will have electoral consequences? Again, Arnold's (1990) work is
helpful. Fundamental to his model is the claim that issues with potential
salience can also influence legislative behavior. When a legislator decides
how to act on a particular piece of legislation, he or she must first determine
whether traceability exists and whether his or her actions on a particular
piece of legislation could become a salient issue in a campaign for reelection.
If so, a legislator must then attempt to judge how a particular action will help
or hurt with particular groups of constituents and whether there are instiga-
tors who can capitalize on his or her behavior. "A talented legislator must
ask himself only two simple questions: First, if I were on the other side,
could Ifigureout a way to incite inattentive publics against me who voted on
the wrong side? Second, are there, in fact, potential instigators who might
mobilize inattentive publics against me?" (Arnold 1990, 70).
In sum, to determine constituent intensity, legislators need to know
whether the effects of a policy are salient enough - or potentially salient
enough - to influence large segments of the electorate. Moreover, legislators
need to know whether there is an instigator who can capitalize on a salient
issue.
If we accept this basic logic of legislative decision making and the informa-
tional needs of legislators outlined in Figure 3.2, how do organized interests
go about trying to influence congressional decision making via an outside
tactic such as stimulating constituent communications? Since there are mul-
tiple steps involved in legislative decision making, and multiple types of
information required, organized interests have multiple options. Once
again, the strategies are indirect but potentially decisive.
Providing Information about the Direction and Intensity of Constituency

Opinion
The first section of this chapter discussed ways in which interest groups
provided information to challengers and voters. Consider now strategies
where interest groups provide information to legislators. One obvious op-
tion is to capitalize on the electoral connection and to provide information
on the electoral consequences of particular courses of action. Grass roots
communications are a particularly effective method of providing legislators
with information on the attitudes and intensities of constituents. Vast
amounts of constituent communications provide essential information to
legislators, constantly trying to gauge how their actions will play out with
constituents. Grass roots communications signal a legislator that a particu-
lar issue is on the radar screen and that constituents are paying attention to
his or her actions. Grass roots communications can raise citizens' - and,
consequently, legislators' - consciousness on an issue.
As Figure 3.2 illustrates, demonstrating that constituent opinion is on
their side is a crucial objective for interest groups. In fact, in some cases, just
showing the direction of constituency opinion can push a legislator in a
group's direction. On many issues, however, other forces are at work and
legislators need to be convinced about the intensity of constituent attitudes
or, in other words, the potential that the issue will become salient in the
minds of important blocks of voters.
Grass roots communications take some of the guesswork out of legisla-
tors' calculations of how their actions could be used against them or whether
there are potential instigators. Stimulated constituent communications en-
able organized interests to prove they have both an argument and the organi-
zational ability to spur constituents to act on a given issue. The information
that groups hope to convey with grass roots communications is that if they
can recruit citizens to write about an issue, they can recruit citizens to vote
on an issue. In short, grass roots communications demonstrate to legislators
that traceability has been established.
Framing Issues and Influencing Opinions

Changing how a vote is labeled or framed is also an effective way for lobbyists
to influence legislators' calculations about the electoral consequences of their
actions. In addition, winning the battle to define what a vote is about can
influence legislative decisions by shaping the actual opinions of constituents
and legislators. Most certainly, how an issue or roll call vote is debated is often
a function of the environment or congressional rules. Still, as discussed in the
section on electoral strategies, lobbyists do not necessarily view the terms of a
legislative debate as exogenous. In fact, a large part of their job is making sure
an issue is debated on terms most favorable to their side.
The highlighting of costs and benefits is an especially important goal of
lobbyists. There are costs and benefits to every bill. Although legislative
entrepreneurs and the political environment prime particular costs and ben-
efits, communicating information to constituents and legislators is a partic-
ularly effective way to stress the consequences that best make an interest
group's argument. In general, it stands to reason that an interest group
would want to tailor its message in the most inviting and effective way. There
is no law that says that the National Rifle Association had to lobby in
opposition to the assault weapon ban with the same language or issue at-
tributes that President Clinton used to support the law.5 Instead of talking
about Uzis and AK-47s, opponents talked about "pork" and midnight
basketball.
Tens of millions of dollars are spent each year researching precisely what
the optimal policy message is for a policy battle, and interest groups have
many possible ways in which to go about framing an issue. Nevertheless,
lobbyists are constrained in their ability to define what an issue is about. No
matter how potentially effective their message may be, they also need a
credible messenger. Grass roots communications can act as such a mes-
senger. An official with an insurance association explained to me, "Newt
and Haley Barbour [RNC Chair] can spin themselves into the ground on
Crossfire, but you need real people to make the spin credible."
Grass roots lobbying can be an effective technique to redefine issues and
influence legislative opinion. Consider the 1993 battle over reduced tax
deductions for business meals. It demonstrated the capacity orchestrated
mass participation has to redefine an issue in more politically acceptable
ways. What policy or electoral reason would a legislator have to support a
narrow provision if it mostly benefits restaurant owners and wealthy mem-
bers of the business community? It is easy to picture a campaign advertise-
ment in which lobbyists are shown dining in a fancy restaurant while a
background voice ponders why Congressman Jones voted to protect a bene-
fit for corporate fat cats.
In part, however, by using the grass roots, the National Restaurant Asso-
ciation (the other NRA) was able to replace this campaign image with an-
other one - one in which a waitress from a small town in Maryland worries
about losing her job and supporting her family if Congress approves this
unfair tax increase. The script of the television advertisement went like this:
"I'm a waitress and a good one. I'd better be because my three sons depend
5
In the same way that using the term "partial birth abortion" advantages pro-lifers, using the
term, "assault weapon ban" defines the issue in a way beneficial to proponents of the ban.
on me. But I might not have a job much longer. President Clinton's eco-
nomic plan cuts business-meal deducibility. That would throw 165,000
people out of work. I need this job. . . . Call 1-800-999-8945 and you'll be
connected to your U.S. Senator" (Dowd 1993).
Using this waitress at press conferences and in television advertisements,
the National Restaurant Association organized food service workers (many
of whom are women and minorities) to contact their members of Congress.
Opposition to the reduction was no longer defending a "fat cat" benefit
but instead was protecting the jobs of hard-working waiters, waitresses,
busboys, and dishwashers (Brinkley 1993a). A tax increase was the message
and working people were the messengers. Utilizing the grass roots, the NRA
was able to demonstrate constituent concern and redefine the issue to its
advantage.
The electoral consequences were reframed using stimulated grass roots
communications. In addition, mobilizing waiters, waitresses, and dish-
washers to communicate with Congress had a ripple effect and changed
general constituent opinion on the issue. In this case, communications from
restaurant workers even helped reframe the issue in the minds of some
liberal Democrats who had previously been for the tax increase. The refram-
ing of the issue also gave those legislators sympathetic to the restaurateurs'
position ammunition to explain what potentially could have been a thorny
vote in favor of special tax loopholes. Legislative entrepreneurs believed they
had defined the issue and the votes in a way to prime populist passions and
group benefits. Instead, in part by using grass roots mobilization, opponents
were able to reframe the legislation as a jobs issue and influence legislative
decision making on a number of different levels.
Summary
Working from an understanding of what influences legislators, I have identi-
fied three basic ways in which organized interests can use the stimulation of
constituent communications as a lobbying strategy. Mobilizing constituent
communications can provide information to legislators on the basic direc-
tion of constituency opinion; can demonstrate that constituent intensity is
high, the issue is salient to important constituents, traceability has been
established, and an instigator exists; and allows interest groups to frame or
reframe issues, influencing general constituency opinion, legislators' es-
timations of how the issue could be used against them, and legislators' prior
policy beliefs. Table 3.2 summarizes the ways in which grass roots lobbying
strategies can influence congressional decisions.6
6
The list of strategies is not meant to be a complete menu of lobbying strategies, but, instead,
Table 3.2. Strategies to Influence Congressional Decisions
Objective Strategy
Influence Congressional Decision Demonstrate the direction of
constituency opinion
Demonstrate the intensity of
constituency opinion - reframe
issue, establish traceability, or
show that instigator exists
Influence constituency opinion by
refraining issue
Influence legislator by redefining
issue
Choosing When, Where, Whom, and How to Mobilize

When the tactic of stimulating constituent communications is chosen, judg-
ments about the weight that an individual's communication to Congress will
have on the ultimate outcome of a legislative fight determine tactical deci-
sions about when, where, whom, and how to target. These judgments are
based on the fact that, unlike votes in an election, communications to Con-
gress are not interchangeable. Comparing voting, the most common and
commonly studied form of participation, to contacting Congress is an excel-
lent way to elucidate this point.
In an election each citizen has one equal vote, and the candidate with the
most votes wins the election in that district or state. 7 The probability of
one's vote influencing the election can be easily calculated. It is identical -
and identically small - for all voters. This, however, is not the case with
communications to Congress. The probability that an individual's com-
munication will influence the process is not necessarily equal to his or her
fellow citizens and is not necessarily small. Instead, it is a function of the
individual communicating, the legislator being communicated to, and the
content of the message.
to include ones in which a grass roots campaign is applicable. For example, Hall and Way-
man (1990) found evidence that mobilizing participation in committee is a strategic objective
of those who donate money to candidates. Grass roots tactics, however, are unlikely to get
legislators to work harder in committee. Unlike PAC donations, which give a member
resources, grass roots lobbying takes time and resources from members. Instead of working
harder, a member who is the target of a grass roots campaign has to allocate precious staff
time and resources to handling the incoming communications.
7
Having grown up in politics in Massachusetts, I learned this much later in life.
For instance, with communications to Congress, not only can individuals
contact Congress as many times as they wish, but the value of their par-
ticipatory acts differs. Bill Gates's ballot for Congress counts the same as a
graduate student's ballot for Congress. However, Bill Gates's letter to his
senator more than likely has a higher weight than a graduate student's letter.
In addition, although votes are anonymous, the identities of those who
contact Congress are known. In fact the identity of the citizen communicat-
ing to Congress is a major piece of information that is being communicated.
Votes are also blunt instruments. They are constrained in their ability to
convey precise information. No matter one's skills, everyone's ballot con-
veys the same type of information. Conversely, communications to Con-
gress convey specific information about the attitudes of particular citizens
on particular issues. Moreover, the value of the information conveyed in the
contact is often a function of the skill and status of the individual crafting the
communication. Finally, there are no fixed rules for tallying communica-
tions to Congress. Although the volume of communications a legislator
receives may be important, legislators do not simply count - or weigh - the
volume of incoming communications and vote the side with the largest
number.
With different and not necessarily small weights for communications to
Congress, it stands to reason that interest groups would want to recruit
those with the highest weights. Given a particular strategy and a particular
type of information that needs to be conveyed, groups striving to maximize
their influence should recruit those with the greatest probability of helping
them achieve their objective. But, with no fixed weights and no fixed rules
for tallying, how do lobbyists go about contacting those with the greatest
probability of helping them achieve their strategic objective? How are
weights calculated and decisions made about when, where, whom, and how
to stimulate constituent communications to Congress?8 What factors influ-
ence the value or weight of a communication to Congress?
With a legislative objective, the essential insight necessary to answer these
questions is that all constituents are not equal and that all legislators are not
equal. The influence of a communication on the policy process is a function
of the individual communicating, the legislator being communicated to, and
the message needing to be communicated. Since individual citizens are
8
Strategic choices also may be restricted by a group's ability to execute the required tactics.
Refraining an issue to change public opinion may be the best strategy, but a group may not
have a credible way to do that. Establishing traceability for swing votes on the Senate Finance
Committee may be the best strategy, but a group may have no members in that state or no
resources to conduct a grass roots lobbying campaign. In addition, answers to the "when,
where, whom, and how" questions of mobilization also depend on each other. Who is
mobilized within a district depends on the member targeted; how a mobilization campaign is
conducted depends on whom groups need to mobilize.
simultaneously recipients and transmitters of strategically targeted elite

pressure, information about the goals of lobbyists and the characteristics of
citizens and legislators needs to be combined in order to calculate a weight.
More specifically, a citizen's probability of influencing the process is a
function of four factors: the probability that an individual citizen's input will
influence a legislator, the probability that a legislator's behavior will influ-
ence the ultimate policy output, the probability that a citizen will participate
if asked, and the probability that a citizen will convey the proper message.
The argument can be represented by the following equation:
W = (IL x OL x IC x PP) x D
where W is an individual's recruitment weight, IL is the influence of a
legislator on the policy process, OL is the openness of a legislator to persua-
sion, IC is the influence of a citizen on that legislator, PP is the probability
that a citizen will contact the legislator if asked, and D is the direction of the
information the citizen will convey. Assuming that interest groups are ra-
tional actors wanting to maximize their influence on the policy process, we
should expect groups to target those citizens with the highest positive
weights. Examining in greater detail what factors influence these proba-
bilities can generate empirically testable predictions about lobbyists' choices
with respect to stimulating communications to Congress.
With an electoral objective, the probability of a citizen's communication
helping a group in the long run by influencing an election outcome is a
function of a set of slightly different factors: a legislator's ideology or posi-
tion vis-a-vis the group, the electoral vulnerability of a legislator, the credi-
bility of a citizen in conveying a message, the probability that a citizen will
participate if asked, and the probability that a citizen will convey the proper
message. It can be represented by the following equation.
W = (ID x V x C x PP) x D
where W is once again an individual's mobilization weight, ID is a legisla-
tor's ideology, V is the electoral vulnerability of a legislator, C is the credibil-
ity, PP is the probability that a citizen will contact the legislator if asked, and
D is the direction of the information the citizen will convey. Once again,
assuming that interest groups are rational actors wanting to maximize their
influence on the policy process, we should expect groups to target those
citizens with the highest positive weights.
Calculating the Influence of a Citizen

Consistent with Fenno's theory about concentric constituencies, lobbyists
know that legislators are likely to give more weight to certain communica-
tions than to others (1978). As one Senate staff member explained frankly,
"There are constituents and then there are constituents." A single letter
from an important member of an important constituency is likely to carry
more weight than a single letter from an unknown member of a less signifi-
cant constituency and, perhaps, even more than many letters.
Depending on their place in a legislator's district view, a letter from a local
leader (banker, Rotary Club president, CEO of a local manufacturing firm)
is likely to carry the most weight. Not only are these types of people more
likely to vote, but they are more likely to be in a position to contribute other
resources and to influence other citizens. Although they represent only three
individuals, few legislators would want to be in conflict with constituents
who have the resources and skills to influence other voters.
Since the value of a contact depends on the strategic objective, there are,
however, exceptions to this rule. For instance, lobbyists hoping to reframe
issues are often better served by recruiting less high-profile constituents. In
the previously discussed case of grass roots lobbying around tax deduc-
tibility of business lunches, the goal was to reframe the issue. Consequently,
to be credible it was necessary for the National Restaurant Association to
mobilize low-income food service workers.
Conventional wisdom has held that "sincere" spontaneous communica-
tions are most likely to influence a legislator. Yet, this view is not consistent
with our understanding of congressional elections, congressional decisions,
and the informational needs of legislators. A communication stimulated by
an interest group may carry more weight than a spontaneous communica-
tion because it carries more information. A stimulated communication may
matter more because a group's organizational ability and intensity are key
pieces of information that are also being relayed.
Calculating the Effect of a Communication on a Legislator

With a legislative objective, the probability that an individual citizen's input
will influence a legislator is also a function of the legislator being contacted.
Characteristics of legislators affect the probability that they will be swayed
by information contained in the communication. A core assumption in my
argument is that the electoral connection matters in legislative decision
making and that communications are an important indicator of intensity
and organizational ability. But I am not claiming that electoral consider-
ations and constituent communications determine congressional behavior
exclusively. Party pressure, previous positions, personal beliefs, and peer
influence are also key components in the decision-making process of legisla-
tors (Kingdon 1989).
Legislators with a long public record in favor of a certain piece of legisla-
tion or who are beholden to powerful committee chairs will be minimally

influenced by communications in opposition to or support of an issue. They
are probably not good targets for groups pursuing a short-term strategy of
influencing legislative votes. Senator Ted Kennedy of Massachusetts, a long
time supporter of national health care, would not have been a good target for
groups mobilizing mass participation around health care reform. Likewise,
Senator Phil Gramm of Texas, who said that the Clintons' health care
reform plan would pass over his "cold, dead political body" would also not
have been a good target. Both men had solid, well-developed positions and
were out of play.
Considering the broader topic of legislative targets, there has been an
enduring debate in the lobbying literature about whether interest groups
should focus on their friends or enemies. Most studies of inside lobbying
hypothesize that groups either try to convert legislators - get them to do
something they ordinarily would not do - or try to mobilize supporters.9 In
my model, tactical targeting decisions depend on the strategies and objec-
tives being pursued. With grass roots tactics being used in pursuit of a
legislative objective, it is the middle that matters. Lobbyists should target
undecided legislators.
However, even with the particular tactic of stimulating constituent com-
munications, other factors are at play and the choice of targets really de-
pends on an interest group's ultimate goal. There are instances in which
interest groups encourage citizens to contact a legislator even when they
know their efforts will have little or no immediate influence on the outcome
of current legislative battles. As discussed previously, messages delivered in
grass roots campaigns are often used to soften up the ground for future
electoral use. Thus, while certain legislators may not be good targets if a
group's goal is simply to win the current policy battle, they may be targeted
so that the issue can be more effectively used against them in a future
electoral campaign.
In this case, the probability of a communication being decisive is a func-
tion of the electoral situation of the legislator and his or her position on
issues of interest to the group. To use their resources most efficiently, we
should expect interest groups to target those legislators in swing districts
where, at the margin, their action on a particular issue has the potential to be
decisive. Similarly, in working to encourage high-quality challengers to take
on a troublesome incumbent, legislators in marginal seats should once again
be the most likely targets.
9
For a particularly intense "discussion" of whether lobbyists concentrate on their friends or
their enemies and what is the best way to study that question, see Austen-Smith and Wright
(1994, 1996) and Baumgartner and Leech (1996a, 1996b).
In short, while other studies have tried to determine whether lobbyists
contact friends or enemies, I argue that tactical targeting choices depend on
the strategic objective and the tactic being used. In a grass roots campaign
with a legislative objective, targets are likely to be undecideds. In a grass
roots campaign with an electoral objective, targets are likely to be persistent
opponents in vulnerable seats.
Calculating a Legislator's Probability of Influencing the Process
For groups pursuing a legislative or vote-influencing strategy, the proba-

bility of a communication being decisive is also a function of various charac-
teristics of the legislator being lobbied. Winning a skirmish with grass roots
communications involves convincing a particular legislator to act in a partic-
ular way. Winning a war with grass roots communications involves getting a
particular piece of legislation passed, blocked, or shaped in a certain way.
From the vantage point of elite lobbyists, the probability of influencing
the process should not be thought of as the probability of a communication
influencing a single legislator to take a particular action, but instead, as the
probability of the communication being decisive in shaping the final legisla-
tive outcome. Depending on the stage of the legislative process, certain
legislators are more important and will be targeted by interest groups. In the
same way that there are constituents and then there are constituents, there
are legislators and then there are legislators. Elite mobilizers do not simply
aim to influence legislators; they aim to influence influential legislators.
A legislator's strategic importance can be judged in many ways. In the
early stages of a legislative battle, influential and persuadable legislators on
key committees should be prime targets for lobbyists. These legislators can
be in positions of committee leadership or can be undecided votes. Early in
the process lobbyists know the crucial battle is whether a bill emerges out of
committee and what shape it takes. Thus, they will concentrate on legisla-
tors who can influence the form of the bill (Romer and Rosenthal 1978).
Certain committees are not only the relevant committees for many types
of important legislation but are also considered to be cue givers to the body
as a whole. We would expect the constituents of members of these commit-
tees to be more likely to be contacted than the constituents of noncommittee
members. Also, as a bill moves to the floors of the various bodies, lobbyists
should focus their attention on undecided legislators and their constituents.
In addition, consistent with Kingdon's findings, certain legislators are cue
givers for their colleagues (1989). Former senator of Georgia Sam Nunn
was an example of a senator who was considered a key cue giver to conserva-
tive southern Democrats on all issues and to all senators on defense issues.
Table 3.3. Objectives, Strategies, and Legislative Targets
Objective Strategy Legislative Targets

Influence Congressional Decision Demonstrate the direction of Undecideds, key committee
constituency opinion members, leadership, cue givers
Demonstrate the intensity of Undecideds, key committee
constituency opinion - establish members, leadership, cue givers
traceability and show that instigator
exists
Influence constituency opinion by Undecideds, key committee
redefining issue members, leadership, cue givers
Influence legislator by redefining Undecideds, key committee
issue members, leadership, cue givers
Influence Congressional Election Define issue and/or establish Opponents in marginal seats
Encourage potential challengers Opponents in marginal seats
Define issue and/or establish Supporters in marginal seats
Discourage potential challengers Supporters in marginal seats
Depending on the issue, lobbyists should therefore try to stimulate lobbying
activity among constituents of key cue givers (Mathews and Stimson 1975).
In Table 3.3 I connect the different strategies to tactical decisions about
where to target.
Calculating and Influencing the Probability That a Citizen Will Contact

Congress
Although there is no registration requirement or legal cost for those who

want to contact their legislators, there is also no official structure for par-
ticipating. Citizens must know more and possess greater skills to communi-
cate with Congress. Prospective contactors must know a debate about an
issue is taking place, grasp some basic details about the debate, know how to
communicate these basic details, and determine whom to contact.
All in all, there is less socialization, less media attention, and no official
structure for participation between elections. Therefore, to encourage par-
ticipation between elections, lobbyists must not only provide an impetus to
participate, but they must often provide the means to participate. No matter
how educated, connected, and efficacious one is, a push is often necessary
to elicit any type of communication to Congress.
Yet, surprisingly, just how recruitment contacts lower participatory costs
has never been examined in detail. Although it has been cursorily examined
in the electoral participation context, the focus has not been on how recruit-
ment actually lowers costs. Instead attention has been paid to how contact-
ing influences individuals' estimations of the benefits they will receive, the
probability that their votes will make a difference, and their perceptions of
the cost of voting. Using the rational choice calculus to describe the par-
ticipation decision, Wielhouwer and Lockerbie argued that "The party con-
tact may raise an individual's estimation of either P or B (or both) increasing
that person's belief in the efficacy of his or her involvement or of the benefits
accruing from the candidate's victory. Or the contact may reduce an individ-
ual's estimation of the costs associated with participation" (1994, 213).
My claim is different. In essence, I am arguing that stimulating participa-
tion between elections not only changes perceptions and provides a nudge
for those already in a state of readiness to act, but it also creates the means to
act. It is conventional wisdom among "get-out-the-vote" (GOTV) profes-
sionals - and has been demonstrated empirically - that mobilization con-
tacts in the electoral context can increase voting rates by modest levels of 2
to 3 percentage points (Goldstein 1994). In terms of stimulating constituent
communications to Congress, we should have much higher expectations
about the effect of a recruitment contact. Contacting Congress is not a
spontaneous activity. Citizens contact their legislators when someone asks

them to and shows them how.
Although elite recruitment is a powerful stimulus, it does not guarantee
participation. Interest groups should also seek to target those citizens who
are most likely to participate if asked. Thus, because they tend to have more
influence, and are more likely to respond to a recruitment request, we
should expect citizens with higher levels of education and stronger connec-
tions to political life to be targeted. In addition, consistent with the claims of
social psychologists, all else being equal, costs will elicit a more intense
reaction in opposition than a benefit should elicit in support (Tversky and
Kahneman 1981). Constituents likely to sustain direct costs as an immedi-
ate consequence of a particular policy should be more likely to contact
Congress if asked and should be targeted by lobbyists.
Moreover, interest groups may not view as purely exogenous the likeli-
hood that a citizen will participate if asked. By crafting their exhortations in
ways that accentuate direct costs, interest groups may increase a citizen's
probability of participating. Finally, lobbyists are not only able to facilitate
participation by lowering the costs of participation, but by raising the costs
of not participating as well. Professional sanctions to those with whom they
have a business relationship and social sanctions to those with whom they
have a social relationship can be brandished. It stands to reason then that
groups would want to target those individuals in social networks and busi-
nesses over whom they have some leverage (Rosenstone and Hansen 1993,
38; Verba et al. 1995a, 139).
Calculating the Probability That the Correct Message Will Be Conveyed

As I have argued, lobbyists want to ensure that their resources are not
wasted on citizens who will not follow through if asked or who will not have a
large influence on the process. In other words, lobbyists want to target those
with the greatest probability of making a difference if asked. Even more
importantly, however, interest groups want to be sure that the citizens they
mobilize will convey the correct message. As one lobbyist told me, "The first
rule is do no harm!" It would be unfortunate for interest groups if they
consistently wasted resources targeting the wrong districts, citizens who
were not likely to influence the process, or citizens who were unlikely to
contact their representative if asked. It would be disastrous for groups if they
activated an influential citizen in the district of an influential legislator who
conveyed information that helped their opponents!
Although the discussion up to this point has concentrated on how interest
groups calculate the absolute value of the probability that an individual will
contact Congress and that his or her contact will make a difference, the sign
of that value is even more crucial. Interest groups want to make sure that the
right message is conveyed, and we should expect reliable allies to be the
targets of mobilization efforts.
Summary and Predictions

Tactical decisions are driven by the desire to frame issues and lower the costs
of participation for those citizens who have the greatest probability of help-
ing a group convey the information most likely to allow them to achieve a
strategic objective. Consider once again the calculus of a mobilization
weight when the strategic objective is a short-term legislative one.
W = (IL x OL x IC x PP) x D
To maximize the value of W, the probability that a contact will achieve a
given strategic objective, we should expect interest groups and lobbyists to
make the following sorts of choices about when, where, and how to recruit.
Lobbyists should target likely allies, citizens with greater resources and
greater influence, citizens in the districts of undecided legislators or key
committee members, and citizens represented by legislators considered cue
givers.
Now consider again the calculus of a mobilization weight when the strate-
gic objective is to influence an upcoming election.
W = (ID x V x C x PP) x D
In this case, W is maximized by a slightly different set of mobilization
choices. Interest groups should target citizens in those districts or states
where a legislator who is at ideological odds with their group's positions is at
risk and those citizens who can most credibly convey the desired message.
No matter the objective, mobilizers should target citizens who are likely to
face early-order, direct costs of legislation as well as citizens in social net-
works or businesses where they have financial or social leverage. Mobilizers
should also frame issues in ways that accentuate early-order, direct costs.
Finally, no matter the strategy, we should expect mobilization contacts to
have a decisive influence on a decision whether or not to communicate with
Congress.
Chapter Summary
This chapter has provided a theoretical framework for investigating why
grass roots communications matter and how they are used as a lobbying
tactic. The structure of congressional elections and legislative decisions,
specifically the need for particular types of information, provides strategic
opportunities for lobbyists to influence the policy process. In the same way
that legislators have multiple motives and thus require multiple pieces of
information, interest groups have multiple ways to provide that information.
Although others have also argued that interest groups strive to communicate
information with their lobbying efforts (Kingdon 1989; Rothenberg 1992;
Smith 1984; Wright 1996), I have detailed the specific kinds of information
that stimulated constituent communications can convey. Moreover, work-
ing from the premise that targeting decisions are driven by the probability
that an individual's communication will influence the process, I examined
the tactical choices of lobbyists and derived a set of testable predictions.
Different from other studies, I also took into account how tactics flow from
strategies and how different strategic objectives demand different tactical
choices.
In the following chapters, I set out to test these hypotheses and the politi-
cal logic of mobilization from which they are derived. I test the predictions
and explore how mobilization choices are made in two different ways. First,
through interviews and observations of selected cases, I investigate the deci-
sions of mobilizers to see if their choices are consistent with my theoretical
framework. Second, I use surveys of the mass public in conjunction with
information on targeted areas to determine who contacts Congress, who is
mobilized, and the effect of mobilization.
Explaining Lobbying Decisions
Well, politics.
Union field director's response when asked what
factors influenced the union's targeting decisions
The three most important aspects of grass roots

lobbying are targeting, targeting, and targeting.
Trade association field director
In the previous chapter I laid out a theoretical framework to explain grass

roots lobbying choices. Using information I gathered on grass roots lobby-
ing campaigns, I now begin the task of investigating if the actual methods
and lobbying choices of interest groups are consistent with this theoretical
framework. The heart of this chapter consists of data gathered from forty-
one formal interviews with interest group representatives. As discussed in
the Introduction, the unit of analysis in this data set is neither the individual
whom I interviewed nor the group he or she represented. Instead, the unit of
analysis is the individual grass roots campaigns and lobbying choices that
were made.
Altogether, the forty-one interviews covered ninety-four separate grass
roots lobbying campaigns across fifteen issue domains. Because of the tim-
ing of my study, campaigns connected with health care compose almost one-
quarter (22 percent) of all lobbying campaigns in my sample and almost
one-third (31 percent) of the lobbying campaigns with a legislative objec-
tive. Since I had sufficient cases and extra information on health care lobby-
ing and so as not to overwhelm all the other issue advocacy campaigns, I
concentrate on health care lobbying decisions alone in Chapter 5. In this
chapter, I focus on the fourteen remaining issues on which I gathered
information.
In the first section of this chapter I examine the strategic objectives of the
various lobbying campaigns contained in my sample. In the second section,
53
I examine the tactical choices of groups in lobbying campaigns with legisla-

tive objectives. In the third section, I describe tactical choices in lobbying
campaigns with electoral objectives.
Strategic Objectives
Tactical choices rely on the strategic objectives of interest groups. Tactical
choices depend on the information groups want to convey with their efforts.
More specifically, with a legislative objective, lobbyists strive to convey in-
formation to members of Congress on the electoral consequences of their
actions. With an electoral objective, lobbyists strive to convey information to
constituents that make it more likely that a particular legislator's action will
be a decisive factor in an upcoming election. In addition, with an electoral
objective, lobbyists seek to convey information to potential challengers that
a legislator is vulnerable and that groups have both a message that resonates
and the ability to deliver it. Again, as discussed in previous chapters, I
differentiate between independent expenditure campaigns or issue ad-
vocacy campaigns that are conducted during the campaign season and those
conducted with more long-term objectives well before the start of the formal
campaign.
Of the fourteen issue areas in my study that attracted grass roots lobbying
campaigns, five had a legislative objective, four had an electoral objective,
and five had mixed strategic objectives. Issues on which a pure electoral
strategy was at work included Medicare and the 1996 budget, worker safety,
term limits, and meat inspection. A pure legislative strategy was in place
with NAFTA, tort reform, and lobbying reform as well as with lobbying
related to telecommunications and tobacco issues. The battle over the Clin-
ton budget and stimulus package in 1993, partial-birth abortion, the crime
bill, campaign finance, and the balanced budget amendment were issues
that drew a mixture of strategies with some groups pursuing electoral objec-
tives and others pursuing legislative objectives.
Table 4.1 reports the distribution of strategic objectives by grass roots
lobbying campaign. An implied electoral threat is clearly a crucial part of a
grass roots campaign with a legislative objective. Nevertheless, corporations
and trade associations in my sample were unlikely to engage in grass roots
lobbying campaigns with direct electoral objectives. Electoral objectives
were more likely to be pursued by ideological groups and unions.
The majority of the seventy-three mobilization campaigns analyzed in this
chapter (60 percent) had a legislative objective. Less than four in ten (36
percent) had an electoral objective. An analysis including health care reform
campaigns would have skewed the distribution even more toward grass roots
campaigns with a legislative objective.
Explaining Lobbying Decisions 55
Table 4.1. Strategic Objective by Number of

Mobilization Campaigns
Objective Frequency
Legislative 44 (60%)
Electoral 26 (36%)
Organizational 3 (4%)
Total 73
As Table 4.1 also shows, there was another possible strategic objective -
organizational maintenance. The groups included in my sample probably
underestimate the amount of grass roots lobbying that goes on in order to
build and maintain an organization. Again, no claim is made that my
sampling methods provide information on the distribution of lobbying
strategies.
Examples of Strategic Objectives

The legislative battle over NAFTA exemplified a lobbying campaign with a
pure short-term legislative strategy. Although the electoral threat was ex-
plicit in union tactics, the immediate goal was to pass or defeat the trade
agreement. Unions sought to convince members of Congress that constitu-
ent opposition was intense - that the issue was important to large blocks of
voters, that traceability had been established, and that an instigator existed.
Interviews with those on the other side of the debate suggested that the
objective of many in the business community was to use constituent com-
munications as a counterweight to the union efforts. In essence, there was
nothing fancy about the debate over NAFTA. Both sides were competing to
see who could convey the most compelling information on the political
consequences of a particular course of action. A woman involved in the
business coalition to support NAFTA explained:1
If NAFTA was a secret ballot, it would not have been close and we would not have
had to spend any time on lobbying. But it obviously wasn't and members who were
inclined to be with us were taking a pounding at home from the unions and the
environmentalists - and from Gephardt and Bonior here in Washington. They
needed an excuse to be with us. They needed some sort of signal that they were not
going to be hung out to dry on this. We even had members asking us to get constitu-
ents to contact them.
1
Unless otherwise noted, all data and results presented in this chapter rely on my interviews.
With the 1993 budget, opponents of President Clinton's plan attempted

to establish traceability on certain elements of the package. The objective
was to uncouple parts of the bill to highlight the short-term direct costs of
the plan - higher taxes - as opposed to the more long-term and indirect
benefit of deficit reduction. In short, opponents of the Clinton package
wanted to frame the vote in Congress as one for higher taxes, while propo-
nents of the plan wanted to frame the vote as one for deficit reduction and
lower interest rates.
Even though the messages they were using were similar, different groups
had different strategic objectives. Some had an electoral strategy in mind,
while others pursued a more short-term strategy to amend or defeat the
Clinton budget. In some cases, individual groups may have pursued differ-
ent strategic objectives for different legislators. In mobilizing constituent
communications to Congress about the Clinton budget, one group in my
sample targeted certain legislators it thought could be persuaded and tar-
geted others because it was an excellent opportunity to prime an issue for
future electoral use.
In addition, the grass roots campaigns conducted during the summer of
1993 around the Clinton budget and stimulus package were also used as a
tool to recruit candidates for the 1994 general election. The director of one
right-leaning ideological group told me, "One year out is a key time for
challengers, and we wanted to convince the good ones that there were issues
they could use and members who would be vulnerable."
Dual strategic objectives were also the case with pro-gun groups in the
summer 1994 fight over the crime bill. These groups targeted some mem-
bers in an effort to sway their votes and targeted others to prime the issue for
use in the impending fall campaign. The crime bill debate also illustrated
how groups can use grass roots communications to redefine issues. For
example, the National Rifle Association may or may not have been outraged
by midnight basketball and dance lessons. The NRA leadership, however,
knew that it would rather debate the relative merits of liberal social programs
than the relative merits of an AK-47. As one lobbyist involved in the lobby-
ing campaign around the crime bill declared: "The first rule of politics is
change the subject." Although the NRA used Second Amendment rhetoric
in its communications to gun owners, the more general strategic objective
was to reframe the crime bill. Accordingly, the message to the general public
was about midnight basketball, dance lessons, and pork.
Congress eventually passed both the crime bill and the Clinton budget,
but by razor-thin margins. Although the Clinton administration and the
Democratic congressional leadership won the legislative battle, they lost the
battle to define these particular issues. In the case of the budget bill, the
legislation was defined by its tax increases. In the case of the crime bill, its
spending components defined the legislation. Moreover, because of the
extremely close margins of victory, every member who voted for the two bills
was susceptible to the charge that he or she cast the deciding vote. In the
months preceding the 1994 elections, then Clinton pollster Stan Greenberg
often talked about running the campaign with a three-legged stool compris-
ing deficit reduction, the crime bill, and health care reform. According to
another Democratic campaign consultant, stimulating constituent com-
munications was one way that Republican groups and their sympathizers
"sawed the legs off Stan's damn stool."
The Clinton White House apparently learned the lessons from the hits it
took on the budget and crime bill battles as well as its defeat in the health
care battle. Using many of the same tactics - commercials, speeches, and
field organizers in key districts - Clinton and his interest group allies went
after elements of the Republican's Contract with America. According to at
least one published account and one deposition, the strategic and tactical
targeting decisions were made right in the White House and were often
directed personally by President Clinton (Woodward 1996).
The strategic objective of Democrats and their interest group allies after
the 1994 election was identical to the strategic objective of Republicans and
their interest group allies after the 1992 election. Groups stimulating con-
stituent communications around the 1995 budget battle and elements of
Newt Gingrich's Contract with America did so to frame the issues and
establish traceability for use in the 1996 elections. The battle was over
whether the GOP budget would be framed as reducing the deficit and saving
Medicare or as slashing Medicare to fund a tax cut for the wealthy. (Of
course, the fact that the two sides were only $30 billion apart over a seven-
year period - less than 1 percent of Medicare spending - influenced neither
the tone nor the political consequences of the debate.)
The not-so-subtle Democratic strategy was designed to put the Republi-
cans on the defensive. The goal was to put Democratic challengers in posi-
tion to use Medicare as a campaign issue as they headed into an election
year. Also, most of these grass roots campaigns were conducted in the fall of
1995 when potential candidates were making decisions about whether to
make runs for Congress in 1996. In fact, officials of three unions with whom
I spoke were explicitly using mobilized constituent communications to Con-
gress as a recruiting device in a few selected districts and states.
One final example illustrates how groups not only have strategic objec-
tives on issues that concern them, but often take advantage of strategic
opportunities when offered. An environmental group in my sample in-
structed its field staff to mobilize constituent communications and create a
stir about GOP efforts to relax meat inspection procedures. This issue was
not really one that was in the group's normal domain, and the targeted
legislators were unlikely to be swayed by the communications. Nevertheless,

this environmental group viewed it as a strategic opportunity to score some
political points and define an issue for future electoral use. The group's field
director put it this way: "Most of our membership are vegetarians and
probably felt like meat-eaters had it coming, but the whole e-coli issue
provided us with a great opportunity to score some hits and soften up a few
of our enemies." The opportunity was especially attractive because many of
the most competitive districts in the 1996 congressional election were in the
Pacific Northwest, where concern about e-coli was high after an outbreak in
a chain of fast-food restaurants.
Summary and Expectations

Interest groups have different strategic objectives when they stimulate con-
stituent communications to Congress. A group's choices about when,
where, whom, and how to mobilize depend on these motives and goals as
well as the type of information it needs to convey. Given the types of objec-
tives discussed here, to maximize the probability that a contact will achieve a
given strategic objective, we should expect interest groups and lobbyists to
make the following sorts of choices.
The forty-four grass roots campaigns with a short-term legislative objec-
tive needed to provide information to members of Congress on the electoral
consequences of a particular course of action. In these campaigns we should
expect to find evidence that interest groups targeted citizens in the districts
or states of key committee members, undecided legislators, and those con-
sidered cue givers. In the twenty-six grass roots campaigns with a long-term
electoral objective, where lobbyists needed to provide information to citi-
zens and potential challengers, we should expect to find evidence that mobi-
lizers targeted citizens in marginal districts or states.
Within these targeted districts or states, mobilizers should have targeted
citizens who were likely to face early-order, direct costs, citizens sympathetic
to their cause, citizens with greater resources and greater influence, and
constituents in social networks or businesses where they wielded financial or
social leverage. Finally, no matter what the objective, the messages used to
mobilize constituent communications should have been framed in ways to
accentuate early-order, direct costs.
Targeting Choices and the Legislative Objective

Targeting is a two-stage process. Lobbyists must decide initially in which
districts and states they will undertake their mobilization efforts and then
whom to contact within those districts and states.
Table 4.2. Timing of Mobilization Campaigns with a

Legislative Objective
Targets Mobilization Campaigns

Committee Only 19 (43%)
Floor Only 10 (23%)
Mixed 15 (34%)
Total 44
When and Where to Target
My first step is to look at the decisions about when and where to employ
grass roots lobbying campaigns. As shown in Table 4.2, my interviews re-
vealed that interest groups with a legislative objective - passing, blocking, or
amending a particular piece of legislation in the short term - were likely to
target swing members on key committees early in the process as well as cue
givers and undecideds when and if the bill came to the floor.
Previous work on lobbying has suggested that committee-based lobbying
is reserved for insiders and has only the subtle objective of getting members
to work harder (Hall and Wayman 1990; Wright 1996, 45). My data indi-
cate that outside strategies such as stimulating constituent communications
are also pursued at the committee level. As Table 4.2 demonstrates, over
four in ten (43 percent) of the grass roots lobbying efforts in my sample were
conducted only at the committee stage. Furthermore, more than three out
of every four (77 percent) of the forty-four grass roots lobbying campaigns
with a legislative objective targeted at least some of their efforts at the
committee level.
Considering individual legislative targets, my model assumes that tactical
targeting decisions depend on the strategies and objectives being pursued.
When the objective is a short-term legislative one, my model predicts that
undecided legislators should be targeted. Consistent with this logic, there
was strong evidence in my interviews that lobbyists did, in fact, target un-
decideds when their strategic objective was to influence short-term legisla-
tive decisions and their tool was to stimulate constituent communications.
Specifically, as Table 4.3 illustrates, more than eight out of every ten grass
roots lobbying campaigns with a legislative objective (82 percent) focused
their efforts primarily on undecided legislators. To reiterate, the claim is not
that undecideds are always targeted but that targets depend on strategic
Table 4.3. Member Targets with a Legislative Objective

Supporters 4 (9%)
Opponents 4 (9%)
Undecideds 36 (82%)
Total 44
objectives and that orchestrated constituent communications are likely to be

directed at undecideds.
When the energy tax was debated, opponents focused their efforts on the
state of Oklahoma, which had three powerful officials in Congress who were
potential swing voters: Senator David Boren, and Representatives Dave
Me Curdy and Bill Brews ter. During the debate over the crime bill, groups
with a legislative objective targeted undecided moderate Republicans from
the Northeast and conservative Democrats from the South and West. On
NAFTA, the targets included twenty-three House members who had not
announced a position on the trade agreement. In addition, pro-NAFTA
forces put field directors in four states they believed would be critical:
Oklahoma, Kansas, Missouri, and Texas. By mid-November, as more and
more legislators took positions and as passage in the Senate seemed safe,
attention was focused on fewer than ten undecided House members. With
the debate over the Hyde amendment on public funding for abortion, tar-
gets included twenty House members who did not have a consistent pro-life
or pro-choice voting record.
There has also been a distinction in the lobbying literature between inside
strategies and outside strategies. According to the lobbyists with whom I
spoke, however, decisions about when and where to pursue an outside
strategy were made very much in tandem with inside lobbyists. Groups
worked with their own lobbyists as well as other allies in both the legislative
and interest group communities. The communications director for a major
insurance company explained: "We work hand in glove with our Hill lobby-
ing staff. That is how we know what the key committees are and who the key
committee members are."
Finally, even though the following declaration requires further empirical
study, more than a few respondents made the point that a major benefit of
the access they get with inside lobbying is to gain political intelligence for use
with other lobbying tactics. One trade association official summed it up as
follows:
If I'm out playing golf with some congressman or I buy a senator lunch, I know I'm
not buying a vote. I'm not even buying access to convince him. What I am buying is
intelligence. He might tell me that he is undecided on some bill that concerns me. He
might tell me about how other members of Congress or fellow committee members
are feeling about an issue. He tells me information that I can then take back and say,
"Hey, we have an opportunity here," or "Hey, we better shore up our support here."
Whom to Target
According to my theory, grass roots campaigns with a legislative objective

should target citizens who have both the greatest probability of influencing a
member of Congress and who are most likely to act in the desired direction.
As discussed already, however, lobbyists target specific members. There-
fore, the first step must be to match potential citizen targets to their correct
political geography. No matter what potential impact a citizen's com-
munication will have on a particular legislator, lobbyists want to insure that
it is directed at a legislator who is likely to be influenced and to influence the
final process.
Advances in district-matching software allow interest groups to identify
the congressional districts of constituents and have played an important role
in the rise of grass roots as a lobbying tool. One company's name in particu-
lar, Legislative Demographic Services (LDS), came up in many of my inter-
views. The firm entered the grass roots lobbying business through its work
with the Republican National Committee during redistricting after the 1990
census. LDS digitized 8,276 federal and state boundaries, defining all the
various political geographies (congressional district, state senate district,
and state house district) in the country by their latitudes and longitudes.
They have also digitized sixty million nine-digit zip codes (zip plus four) and
matched them to their correct latitudes and longitudes. For those addresses
that only have a five-digit zip code available, they use another set of software
that matches street addresses to their correct nine-digit zip codes. In this
way, with a high degree of accuracy, the company is able to match members,
employees, stockholders, vendors, or lists of potential sympathizers to their
correct legislative districts.
Judging from the information I gathered, once individual targets were
matched to the correct district, lobbyists pursuing a legislative objective
were likely to concentrate on citizens who had the greatest probability of
influencing a legislator - targets that they typically dubbed "key contacts" or
the "grass tops." As Table 4.4 illustrates, pure broad-based campaigns were
utilized in fewer than one in five cases when groups were pursuing a legisla-
tive objective. Although broader-based grass roots lobbying campaigns were
used in conjunction with more focused efforts about a third of the time,
Table 4.4. Constituent Mobilization Targets with a

Mobilization Targets Mobilization Campaigns

Broad Based 8 (18%)
Key Contacts 21 (48%)
Mixed 15 (34%)
Total 44
almost half of the lobbying campaigns in my sample concentrated solely on

higher-end constituents. As I discuss shortly, however, there are exceptions.
Consistent with these findings, the interviews also pointed to the tremen-
dous amount of resources directed at identifying those who have a high
probability of influencing a legislator. One company surveyed its employees
and stockholders to determine the extent and nature of existing political
connections. The company's lobbying chief explained her reasoning: "The
congressman may be a brother-in-law or sister-in-law of an employee. He
may be a neighbor, a fellow church deacon, or, say, a fellow club member,
[or] have his kid on the little league team. We need to know that kind of
stuff."
Often, such tactics are described as mobilizing the "grass tops." Jack
Bonner, one of the most prominent consultants in the grass roots lobbying
game specializes in recruiting such high-profile constituents. For getting a
community leader to communicate his or her feelings to a legislator on the
phone or in writing, Bonner will charge from $350 to $500. For setting up a
personal meeting between a high-profile constituent and a member of Con-
gress, Bonner can charge as much as $9,000 (Stone 1993).
A lobbying director for a trade association described to me a software
program she used that searches for personal and professional connections to
members of Congress. The software was originally developed by a company
working for the Securities and Exchange Commission to track insider trad-
ing. The software looks at multiple sources - club memberships, university
alumni lists, and fraternity and sorority lists - to determine who might have a
relationship with a member of Congress.
More commonly, groups matched up lists of political contributors and
supporters to their rosters of employees and stockholders. A manual by the
chemical industry on how to set up a grass roots operation recommends:
A Washington Representative should have a master list of people in the congressional
district or state where his company has operations, and know who has supported
members, politically and otherwise, and who has some influence. Those who have
supported members financially and/or helped sponsor fund-raising activities are in a
better position to request support than those who haven't. Members of Congress
react to key people in their districts and states, through leaders and financial backers,
and it is up to the Washington representative/lobbyist to determine who can best
approach a member.
Consistent with expectations, groups do expend significant resources
determining who has a high probability of influencing a legislator. I also
learned that they do not necessarily take this probability as fixed. In other
words, they also work to increase the probability that a communication will
be decisive. Quoting from this same chemical industry manual: "The Wash-
ington lobbyist must urge field personnel, from regional vice presidents to
local representatives, to get to know their Congressmen and Senators and
become well-acquainted with Members' district personnel." A manual from
a pharmaceutical company recommended that local managers "raise their
political visibility" by delivering in person their firm's PAC checks.
In addition, the goal of grass roots lobbying campaigns is not only to
encourage influential constituents to contact influential legislators, but as
one corporate official put it, "To get people to contact Congress in a way
consistent with your company's beliefs." Organizations want to be sure that
they are not mobilizing employees who could convey the wrong message.
For example, in the battle over NAFTA, two telephone companies actively
stimulated constituent communications to Congress. They did not, how-
ever, recruit any of their employees who were members of the communica-
tions workers union.
Leverage
Lobbyists not only want to mobilize those who will communicate the right
message and who are influential, but they also want to make sure that the
targeted citizens follow through on their requests. To this end, they often
work with those over whom they enjoy social or financial leverage. As Table
4.5 demonstrates, members and stockholders as well as employees and
vendors were most likely to be targeted by interest groups to contact
Congress.
Many corporations have formalized their employees' public affairs re-
sponsibilities. For instance, one company I studied expected its plant man-
gers to contact their legislators four times a year and conduct at least one
plant visit for their member of Congress. Political activities are a part of each
manager's own performance review. Said one corporate public affairs offi-
cial, "We have some legal problems if we put lobbying in an employee's job
description, but we make it known what we expect. We give people mone-
Table 4.5. Constituent Mobilization Targets with a

Mobilization Targets Mobilization Campaigns

Members/Stockholders 19 (43%)
Employees/Vendors 11 (25%)
Third Party 7 (16%)
Mixed 7 (16%)
Total 44
tary rewards and recognition for their political work." A major insurance
company "strongly suggested" to its midlevel executives that they take a
three-day course on legislative lobbying and grass roots tactics. During the
debate over NAFTA, automakers worked through their suppliers to stimu-
late letters and phone calls in support of the agreement. In particular. Ford
Motor Company worked through its five thousand dealerships (Stone
1994).
Ideological groups and unions also target those over whom they have
financial or social leverage. A union representative explained how his union
uses "peer pressure" to get people involved in politics. "People see each
other at work every day and we want to make people embarrassed if they
have not done their duty." The representative of a group that concentrates
its work in the churches explained that "seeing your fellow parishioners
taking part and doing so publicly creates a real pressure to join in."2
Finding Allies
Although the preceding data and discussion reveal a concentration on key
contacts and people over whom groups had leverage, the data also show a
significant level of targeting of nonmembers or nonemployees. One consul-
tant talked about a third-party campaign he had orchestrated. The client
was a major defense contractor trying to get Congress to approve the build-
ing of a new aircraft carrier. To determine where it had leverage, the first
thing the consultant did was to study the firm's accounting lists of suppliers.
He then set up a conference call with 425 of the suppliers and informed
them that they would be expected to have their employees write letters,
make phone calls, and conduct plant tours for selected members of Con-
2
See Chong (1991) for an excellent study of internal group pressures and participation in the
civil rights movement.
gress. He then worked with the firm's Capitol Hill lobbyists to come up with
a list of eighty senators and representatives they needed to convince. Out of
the 425 suppliers, 300 had locations in the targeted districts and states.
These companies were then sent precise marching orders and materials on
how to initiate letter-writing and phone campaigns. Included with the
marching orders and materials was a memo outlining how much the defense
contractor had spent with the supplier and what would be spent in the
future. The not-so-subtle message was that their future business relation-
ship would hinge on their enthusiastic participation in the mobilization
effort.
Also, there was evidence of broader-based mobilization efforts. One trade
association representative explained simply, "If you don't have credible mes-
sengers, you need to develop allies who will take action." Her point made
the argument that a citizen with a high probability of influencing the process
is not always wealthy or well connected. Credibility is also important. For
example, in a previous chapter, I discussed how the auto companies hired
Jack Bonner to find credible allies. In that case Bonner recruited seniors and
Boy Scout representatives to contact Congress in opposition to provisions in
the Clean Air Act. Although there surely was a place for the CEO's of Ford,
Chrysler, and General Motors, the objective of the grass roots strategy was
to provide information that was best conveyed by average people. In such
cases, communications can also have a ripple effect.
The battle over tort reform presents a similar case. Insurance companies
could not get out in front of the effort to change liability laws. Therefore,
instead of mobilizing just insurance salesmen, they targeted educators,
little-league coaches, and small-business owners who could convey informa-
tion to legislators on their frustration over the explosion of frivolous lawsuits
and the cost of liability insurance.
Groups also use different messages to stimulate different target audiences
to contact Congress. A tobacco company felt that it needed more - and
more credible - citizens contacting Congress on the topic of whether the
Food and Drug Administration should regulate cigarette manufacturing
and sales. The company's government affairs staff came up with a three-part
strategy. First, they mobilized smokers with the "stay out of my life" mes-
sage. Second, they mobilized small stores worried about more taxes and
regulations. Third, and most creatively, they targeted gay-rights groups with
the message that the FDA's preoccupation with tobacco was distracting
them from approving new drugs for AIDS. The effectiveness of these efforts
can be debated. Still, this example is illustrative of how lobbyists do not
always target the most connected or the most wealthy to contact Congress.
When pursuing more broad-based or third-party targets, groups obvi-
ously want to make sure that targeted citizens will follow through and con-
tact Congress and that they will communicate the correct message. Many of
the interest group representatives with whom I spoke argued that previous
behavior is the best measure of how citizens will react to requests for par-
ticipation. Consequently, much of their energy was directed at acquiring
information about people's attitudes and past actions.
One of the most successful consulting firms in the grass roots business is
the same company that pioneered direct marketing of credit cards. It spe-
cializes in integrating massive amounts of data from different political, con-
sumer, and demographic sources in order to find the most fertile and
friendly targets for mobilization. If time and resources permit, groups will
conduct surveys off these lists to determine which messages work the best
and what type of political or demographic profile is most likely to respond to
a mobilization request. (Obviously, these "Cadillac" methods cost lots of
money and are used mainly by corporations and trade associations and not
by citizen groups.) As one trade association official explained: "First, we
research who our target audience should be. Let's say it's college-educated
women. We then gather lists of college-educated women. We test what
messages work best with college-educated women. We buy advertisements
on shows that ratings books tell us college-educated women are more likely
to watch."
Technologies and Making Contacting Easy

New technologies are used by groups and consultants to make contacting
Congress as easy as possible and to lower the costs of participation. In the
1996 fight over Medicare, a union representative explained to me how his
organization sent field staff to construction sites with cellular phones.
"Patch-thrus" are also an extremely popular new technology. With this
method, citizens are called and read a script about a pending piece of legisla-
tion. If they are sufficiently impressed to want to contact their legislator, the
caller immediately forwards their call to the office of their representative or
senator. The calls and the response rates are constantly monitored, and the
most effective messages are honed.
The United States Chamber of Commerce has built an elaborate compu-
ter system and phone bank with the capability of contacting a group's mem-
bership when an issue of concern to the chamber comes up in Congress.
The software is able to sort members according to their geographical loca-
tion and business type. When an issue arises that is of concern to the cham-
ber, the system is programmed to telephone selected chamber members in
key districts. A computer-generated voice informs the contact of the upcom-
ing congressional vote and offers three choices: pressing " 1" sends a letter in
the contact's name to his or her representative; pressing "2" sends a voice
mail to his or her representative; and pressing " 3 " immediately patches the
contact through to his or her representative's office.
A main tactic for the Christian Coalition is to use churches as a way to
construct target lists. First, staff members gather church membership lists
and match names to the correct political geography. Second, they match
these lists to voter registration files to determine party registration and past
voting behavior. Third, they conduct surveys to gauge more accurately the
ideologies of churchgoers and what issues most concern them. Then, when
an issue arises and they want to mobilize communications, they have data on
participation patterns, ideology, and issue concerns. Using these augmented
lists helps them insure that a request will pay dividends and that participants
will communicate a message consistent with the group's beliefs.
The Christian Coalition also makes use of broader-based tactics. During
the battle over the Clinton budget, it aired television advertisements in the
districts of undecided Democrats. The message asked if viewers were sick of
high taxes and invited them to call an 800 number. Software at the location
where the numbers came in immediately matched the telephone number to
the correct congressional district. A packet of materials customized to the
state and congressional district of the caller was sent out the next day. Even
more importantly, the caller's name was then put on a list for future
campaigns.
Tactical Choices and the Electoral Objective

National political tides and the partisanship of a district have a significant
influence on congressional election outcomes. They are also largely immune
from interest group influence. Nevertheless, two factors that can be influ-
enced by interest groups - the quality of challengers, and the way an incum-
bent's record is viewed, also affect congressional elections. As I discussed in
greater detail in the previous chapter, grass roots lobbying is an effective way
for interest groups to provide information to both the electorate and to
potential challengers. More specifically, it is a way for groups to convey
information that may increase the probability that an incumbent's voting
record will be salient and that a troublesome incumbent will face a strong
challenger.
As many of my informants said, for a message to resonate in the campaign
season, it needs to be framed in the legislative season. A field director for a
business association involved in grass roots lobbying activities echoed these
points and talked more about the process of creating defining moments:
"Even if our letter-writing or phone campaign does not change a member's
Table 4.6. Members Targets with an Electoral Objective

Supporters 2 (8%)
Opponents 17 (65%)
Undecideds 7 (27%)
Total 26
vote, it can soften things up for the future. It helps divert a member from
their script, unnerves them, makes them say something stupid that can be
used later."
Where to Mobilize
Consistent with the expectations outlined in Chapter 3, when the strategic
objective was electoral and the goal was to provide information to constitu-
ents and potential challengers, interest groups in my study targeted legisla-
tive opponents. As Table 4.6 demonstrates, about two out of every three
grass roots lobbying campaigns with an electoral objective focused on legis-
lators considered by interest groups to be opponents.
Just being an opponent, however, was not sufficient to draw the attention
of grass roots lobbyists. Groups, in general, only targeted those members
whom they viewed as being potentially vulnerable and who were likely to be
affected by their efforts. Vulnerability was often gauged by a legislator's
previous winning percentage and the partisanship of the district. In 1993
and 1994, Republican-leaning groups targeted Democrats who had sup-
ported Clinton's budget bill and who represented districts won by George
Bush in 1988 or 1992. In 1995 and 1996, the American Association of
Retired Persons, unions, and the Democratic National Committee carefully
targeted their grass roots lobbying efforts around Medicare at Republican
freshmen who won by less than 5 percentage points in 1994 and who had a
large proportion of senior citizens in their districts.
Informal conversations with a number of pollsters on the Democratic side
indicate that the multimillion-dollar effort by unions in 1995 and 1996 to
regain control of Congress for the Democrats also made extensive use of
survey research to determine where to target resources and on which issues
to mobilize. Surveys were taken to determine which Republicans were most
vulnerable, as well as which issues voters found the most important and
Table 4.7. Constituent Mobilization Targets with an Electoral Objective
Mobilization Targets Number of Campaigns

Members/Stockholders 3 (12%)
Employees/Vendors 3 (12%)
Third Party 14 (54%)
Mixed 6 (23%)
Total 26
were the most upset about. In short, the districts that were likely to draw the
fire of the Democratic National Committee and its interest group allies were
the districts where a member was judged vulnerable and the population
showed concern about potential Medicare cuts, changes in worker safety
rules, or environmental regulations.
Whom to Target
As expected, with an electoral objective, citizen targets were less likely to be

members, employees, stockholders, or suppliers and were more likely to be
potential allies from the constituency as a whole. Even large membership
organizations like unions went outside their immediate family. One union
official explained, "Frankly, there aren't as many union members anymore
and lots of them might vote Republican. Seniors are good targets and Medi-
care is a good message, so we're going with that."
As Table 4.7 demonstrates, with an electoral objective, constituent targets
were less likely to be members and stockholders or employees and vendors.
Instead, with an electoral objective, in over half the grass roots campaigns I
examined, interest groups targeted constituents outside of their immediate
memberships or corporate families.
Furthermore, with a long-term electoral strategy, groups were more likely
to pursue broad-based targets and not just go after the most influential
constituents. In other words (and not surprisingly), with an electoral objec-
tive the value of the probability that one's communication will be decisive
was similar to the one-person, one-vote rule of elections. As shown in Table
4.8, in more than two out of every three of the mobilization campaigns with
an electoral objective, interest groups had broader-based constituent
targets.
Table 4.8. Constituent Mobilization Targets with an Electoral

Objective
Mobilization Targets Number of Campaigns

Broad Based 18(69%)
Key Contacts 3 (12%)
Mixed 5 (19%)
Total 26
Examples
The following examples illustrate with actual cases both the types of mes-
sages that were used and the targeting decisions that were made by groups to
stimulate communications to Congress and frame issues before the start of
the formal campaign in 1996.
A commercial produced by the AFL-CIO was aired in the districts of
twenty-seven Republican first-termers who represented districts won by
Dukakis in 1988 or Clinton in 1992. The advertisement was aired during
the summer of 1995 featuring Ron Hayes, an Ohio man whose son was
killed in a work-related accident.
[Man is shown sitting next to his wife]
Hayes: Our son Patrick was a good kid. A real hard worker. Two years ago Patrick
was crushed to death in the feed mill where he worked. The company thought they
could get away with breaking the law. Patrick was just nineteen years old when he
died. Now the Republicans in Congress are cutting health and safety protections I
know can save lives. If they succeed, more people will die.
[Chyron graphic of House member's name and phone number appears on screen]
Narrator: And they've voted to cut Medicare. And college loans. To pay for huge tax
breaks for the rich. Tell Republican Congressman [name] to stop cutting health
and safety so other families don't lose their loved ones. Tell Congressman [name]
to stop attacking America's families. (AFL-CIO release, August 21, 1995)
Later that fall, as part of the Medicare campaign, organized labor tar-
geted twenty-two GOP members representing swing districts and represent-
ing a high proportion of senior citizens. Even though it is impossible to verify
their claims, a union official reported that they made over 500,000 phone
calls in those twenty-two districts. Consistent with this account, the National
Journal reported on December 1, 1995 that elderly groups targeted twenty-
six House Republicans who won with less than 55 percent of the vote and
whose districts had over 12 percent senior citizens.
Chapter Summary
The premise of this chapter was a simple one: to understand how lobbyists
make lobbying choices, one must look at the actual lobbying choices of
interest groups. Using data gathered from seventy-three different grass roots
lobbying campaigns as well as a good bit of soaking and poking, I tested the
theory of targeting choices that was outlined in Chapter 3. Although diffi-
culties in defining and sampling a population of groups pursuing grass roots
lobbying tactics certainly do not make this the last word on tactical choices,
there was strong and consistent evidence that there is little unsystematic or
transitory about the targeting and recruitment choices of political elites.
As conversation after conversation and decision after decision indicated,
interest groups strive to deploy their resources in the most efficient ways
possible. Furthermore, the methods and technologies that I discussed in
this chapter illustrate the premium placed on devising the correct messages,
finding the correct messengers, and identifying those most likely to influ-
ence the legislative process in the desired direction.
These messages and targets were a function of different strategic objec-
tives. There was strong evidence that the need to convey different types of
information to different audiences yielded different legislative and constitu-
ent targets. With a short-term legislative objective, organized interests tar-
geted undecided legislators and the constituents most likely to influence
those legislators. While these were often wealthy or well-connected constitu-
ents, the data also showed that credibility - and creativity - occasionally
demanded nontraditional messengers. When the goal was to provide infor-
mation to voters and potential challengers, targets were likely to be legisla-
tive opponents who were judged vulnerable. Also, with an electoral objec-
tive, the constituent targets were likely to be more broad-based.
In sum, this chapter has provided an overview of how political elites
decide when, where, whom, and how to target. In the following chapter, I
examine the same questions, but do so by taking a more detailed look at one
particular issue - the momentous 1993-1994 battle over the Clinton health
care plan.
Lobbying Decisions and the Health Care
Reform Battle
Step one, decide who will be the key votes on the

key committees charged with health care issues.
Step two, mobilize small business owners who are
influential in their states and districts and are will-
ing to deliver our message. Step three, take the
people from step two and aim them at the people
from step one.
John Motley, National Federation of
Independent Business1
If you don't want government gatekeepers telling

you what doctor you can see, call Congressman
Payne at (202) 225-4711 and tell him to vote "no"
on the Clinton health care plan. That's (202)
225-4711.
Citizens for a Sound Economy,
radio advertisement
Five days after his inauguration as the forty-second president of the United
States, Bill Clinton appointed his wife, Hillary Rodham Clinton, to head a
task force responsible for drafting national health care reform legislation.
Clinton was making good on a campaign promise to tackle an issue that had
risen to national political prominence in the wake of Harris Wofford's upset
victory in Pennsylvania's 1991 special election for the U.S. Senate. Clinton
pledged that the task force would finish its work and that he would submit
his plan to Congress within one hundred days. Due, however, to a more
difficult than expected battle over his budget plan as well as various scandals
and staff mistakes, Clinton's plan was not ready until late summer.
Still, even with the delay, when President Clinton finally presented his
plan to the nation in an address to a joint session of Congress in late Septem-
1
As quoted in Toner (1994).
72
The Health Care Reform Battle 73
ber of 1993, the response was extremely positive. A Gallup Survey on Sep-
tember 24, 1993, conducted in the days immediately following the speech,
reported that almost six in ten Americans (59 percent) supported the plan
and that only one in three opposed it (33 percent). Other public and private
surveys conducted at the same time manifested similar levels of support for
the "Clinton plan." Yankelovich's research had 57 percent of Americans
supporting the plan with 31 percent in opposition. The ABC News/Wash-
ington Post poll had it at 67 percent to 20 percent. Public Opinion Strat-
egies, the pollster for the Health Insurance Association of American
(HIAA), did not include leaners and showed that Americans supported the
plan by a margin of 35 percent to 21 percent.
In the wake of the president's speech, Republican members of Congress
were talking favorably about reform. Even the United States Chamber of
Commerce was saying good things about the Clinton plan. After the first
lady won rave reviews for her performance during initial hearings on the
legislation, support among both the mass public and key legislators even
seemed to rise (Clymer 1993; Martin 1995; Skocpol 1996). In short, during
the fall of 1993, the passage of at least some type of health care legislation
seemed inevitable.
This atmosphere stands in sharp contrast to the state of affairs less than
one year later. On September 23, 1994, House Speaker Tom Foley, House
Majority Leader Dick Gephardt, and Senate Majority Leader George
Mitchell emerged from a meeting at the White House. The trio declared to
the assembled press corps that both houses of Congress were hopelessly
deadlocked and that health care reform was dead in the 103rd Congress.
What happened?
Because of the intensity of the lobbying push surrounding health care
legislation, many claims have been made about the potency of interest group
efforts in killing the Clinton plan and thwarting passage of any sort of reform
in the 103rd Congress. Some observers have pointed to the battle over
health care reform as a typical case of lobbyists controlling the legislative
process and determining legislative outputs (Johnson and Broder 1996;
Lewis 1994; Skocpol 1996). In particular, much attention has focused on
grass roots lobbying and the hundreds of groups that mobilized millions of
Americans to phone, fax, telegraph, mail, and e-mail their representatives in
Washington.
Ira Magaziner, the architect of the Clinton plan, asserted to me in an
interview that "the most effective tactic against our program was grass roots
mobilization and phone banks in swing districts." Dick Gephardt, com-
menting on the effect of various health care lobbying strategies, opined: "It's
not money. It's votes. The common view is that all of these interests came in
and intimidated people by either giving them money or not giving them
money. I think money had little to do with the outcome. It's the political
work they did at home" (as quoted in Johnson and Broder 1996, 195).
The grass roots lobbying efforts of the HIAA and the National Federation
of Independent Business (NFIB) have drawn the particularly close attention
of numerous media observers. The "Harry and Louise" series of television
commercials produced by the HIAA became legendary in media and con-
sulting circles and were credited by many - most prominently Hillary
Clinton - with turning the tide against the Clinton plan. Meanwhile, be-
cause of its successful campaign to strike any sort of employer mandate from
health care reform legislation, scores of pundits crowned NFIB the new
interest group power on the block (Duncan 1994; Headen 1994; Johnson
and Broder 1996; Lewis 1994; Scarlett 1994; Toner 1994; Weisskopf
1994a, 1994b). In fact, in typical inside-the-beltway fashion, the NFIB sent
journalists reprints of many of the articles that claimed they killed the Clin-
ton plan. These journalists then wrote even more articles about NFIB
influence - which were then sent to yet more reporters.
Lobbying efforts in general, and grass roots tactics in particular, alone
cannot explain the demise of the Clinton plan. Some of the credit - or
blame - must go to an overreaching plan devised in secret, a divided Demo-
cratic Party, a determined Republican opposition, and a problem-plagued
presidency.
My goal in this chapter is neither to assess credit or blame nor to gauge the
exact impact of interest group activities and tactics. Instead, the battle over
health care reform during the first two years of the Clinton administration
can teach us much about how lobbyists use grass roots lobbying and why
many citizens participated in politics during the 103rd Congress. Whether
or not it was "typical" or "the most extensive use of grass roots lobbying
ever," there were certainly intense interest group activity and intense mobili-
zation efforts surrounding health care reform. Again, conditions - the sali-
ence of the issue and the volume of lobbying - may have been unique, but
the strategic and tactical choices made by interest groups were not. Health
care reform provides a good case study in which to test the empirical im-
plications of my theory of mobilization choices.
The Data
A subset of the interviews I conducted with interest group representatives
constitutes the major source of data for this chapter. More specifically, I use
information from twenty-one separate formal interviews with representa-
tives of interest groups who conducted grass roots lobbying during the battle
over health care. The interest group representatives with whom I spoke did
not comprise a random sample of all groups mobilizing constituent com-
Table 5.1. Interviews by Type of Group
Type of Organization Pro-Clinton Plan Anti-Clinton Plan Total

Trade Association 1 7 8
Corporations 0 3 3
Left Ideological 7 0 7
Right Ideological 0 3 3
Total 8 13 21
munications around health care - and certainly overstate the big players.
The respondents, nonetheless, offer a good mix of groups by type of organi-
zation, ideology, and their position on the Clinton plan. Table 5.1 reports
the kinds of groups contacted and their positions on the Clinton plan.
To obtain additional background information, I also spoke with Ira Maga-
ziner, the architect of the Clinton plan. For the view from Capitol Hill, I
spoke with two members of the Congressional leadership - one Republican
and one Democrat - who were deeply involved in the debate. I also spoke
with four veteran Washington journalists who covered the health care
debate - David Broder of the Washington Post, Cokie Roberts of ABC News,
Steve Roberts of US News and World Report, and Marty Plisner of CBS
News. My observations as a political consultant during the eight months
preceding the 1994 election also helped to inform my analysis and place
mobilization choices into their correct political context. (See Appendix C
for a chronology of the health care battle and a summary of the major plans
and events.) All in all, the information I was able to gather provides a
relatively broad and deep assessment of the lobbying choices made in the
battle over health care.
Strategic Objectives and Tactical Expectations

The main strategic objective of grass roots lobbying on health care reform
was to influence legislative decisions. Although the battle over health care
reform also provided an opportunity for groups to engage in some organiza-
tional maintenance or to define an issue for future electoral use, the primary
objective of nineteen out of the twenty-one groups in my sample was to pass,
block, or amend the Clinton bill. This is not to suggest that these groups did
not have other goals and did not use other strategies and tactics. In fact, and
not surprisingly, many of the groups in my study used a wide variety of
tactics to achieve a wide variety of objectives. Personal meetings were con-
ducted. Testimony was given. Dollars were contributed.
Since spending on health care comprises one-seventh of America's gross

domestic product, the stakes were enormous. Typical comments from my
interviews included: "This was a matter of life and death for our members."
"We couldn't afford to screw around, the very survival of our industry was at
stake." "We needed to kill this animal now."2 So, given the high stakes and
the immediate goal of influencing a legislative decision, how did groups
employ their grass roots lobbying efforts? What type of information did
interest groups hope to convey by pursuing an outside lobbying strategy
that, in large part, worked to stimulate constituent communications to
Congress?
Because health care was a high profile issue and legislators were facing
many competing pressures, interest groups could not hope to win the battle
by simply providing legislators with information on the basic attitudes of
their constituents. Legislators were already receiving massive amounts of
information from the White House, party leaders, think tanks, and the
press. The challenge for interest groups pursuing a grass roots strategy was
to provide politically useful information. More specifically, the goal was to
provide legislators with information on the potential electoral consequences
of their actions. One corporate lobbyist noted, "They had lots of people
telling them lots of things and giving them lots of information. We had to
provide them with the information that they really cared about, how this
affected their bottom line, how this affected whether they were going to get
reelected or not."
Simply put, legislators needed to know whether they would be held ac-
countable for passing or defeating the proposed health care reform. In Ar-
nold's terminology (1990), health care legislation depended on legislators'
calculations about whether there would be traceability for enacting or not
enacting health care reform. Although the political and economic environ-
ment as well as the skill of legislative entrepreneurs clearly would influence
the nature of traceability on health care, interests groups also worked to have
a say. Grass roots lobbying and stimulating constituent communications
provided one way for groups to have that say.
One of the architects of organized labor's campaign for the Clinton plan
summed it up quite well: "We had to tell our people about the benefits and
get them to tell their congressmen about those benefits before the other side
could tell them about the costs." A business opponent of the Clinton plan
remarked: "We had to explain to members of Congress - through their
constituents - what the Clinton health plan was. We couldn't let the White
House do all the explaining." Another opponent remarked: "First thing we
2
As was the case in Chapter 4, unattributed quotations come from my interviews, which were
done on a not-for-direct-attribution basis. In some cases, when a respondent gave me specific
permission, the source of the quotation is noted.
needed to do was change the subject. If the debate had been about providing
health insurance to middle-class Americans we would have lost. Instead, we
made it about cutting jobs and reducing choice."
These comments were typical of the ones offered in many of my inter-
views. Broadly speaking, the goal of supporters of the Clinton plan was to
focus attention on the general benefit of guaranteed coverage for all Ameri-
cans and to demonstrate the electoral consequences to legislators if no
health reform was passed and if certain elements were not included. The
goal of Clinton plan opponents was to focus attention on the many particu-
lar costs that achieving guaranteed coverage would require and to demons-
trate the electoral consequences to legislators if certain elements of health
care reform were enacted into law.
The agenda was essentially set by the Clinton plan. Lobbying attention on
health care reform quickly focused on four basic areas: employer mandates,
price controls, mandatory alliances, and physician choice. The challenge for
groups concerned with each of these individual issues was to demonstrate
that there was a constituency for their area of concern. In other words, the
goal was to show traceability for a particular element of the health care
debate.
Some scholars have assumed that with a legislative objective as such,
changing national public opinion or getting people from all over the country
to contact Congress was the goal of outside strategies in the health care
battle. Therefore, when they find relatively little national recall for advertis-
ing campaigns like "Harry and Louise" or a weak relationship between
viewership and negative opinions of the Clinton plan, they question the
effectiveness of such outside strategies (Jamieson 1994; West, Heith, and
Goodwin 1995). If changing national public opinion and getting people
from all over the country to contact Congress were not the objectives of
these lobbying campaigns, however, then the fact that such campaigns failed
to do so misses the point about such interest group behavior. "We were
trying to move congressional votes, not Gallup numbers," was how one
lobbyists put it to me. Another explained, "We didn't need to convince
all Americans or all congressmen, we just had to convince the ones that
mattered."
At bottom, convincing the ones that matter is essentially what my theory
of grass roots lobbying decisions boils down to. The theory predicts that elite
recruitment choices should be driven by the desire to frame issues and lower
the costs of participation for those citizens most likely to influence the
ultimate legislative outcome. These citizens are the ones who have the great-
est probability of helping a group convey the information most likely to allow
them to achieve a strategic objective. More specifically, this theory predicts
that groups striving to maximize their influence on legislative decisions
Table 5.2. Where and When Groups Mobilized

Communications
Targets Number of Groups

Committee Only 14 (66%)
Floor Only 0 (0%)
Mixed 7 (33%)
Total 21
should target influential citizens in the districts of those members who are
open to persuasion and who are most likely to influence the legislative
process.
Examining the case of health care reform, we should find evidence that
interest groups stimulating constituent communications to Congress had
targeted citizens in the districts or states of key committee members, un-
decided legislators, and those considered to be cue givers. Within these
targeted districts or states, interest groups should have targeted citizens
sympathetic to their cause, citizens with greater resources and greater influ-
ence, constituents in social networks or businesses where they wield finan-
cial or social leverage, and citizens who were the most likely to face early-
order, direct costs. Moreover, the messages used to mobilize constituent
communications should have been framed in ways to accentuate such early-
order direct costs.
An Overview of Tactical Mobilization Choices

As Table 5.2 illustrates, all twenty-one of the groups in my sample reported
that they mobilized constituents in the districts of members of committees
that were charged with tackling the Clinton bill. In addition, seven of the
groups also mobilized on the floor around the leadership bills that emerged
in the summer of 1994.
Five committees had jurisdiction over health care legislation: Senate Fi-
nance, Senate Labor and Human Resources, House Ways and Means,
House Energy and Commerce, and House Education and Labor. My theory
predicts the grass roots lobbying should have been pursued in those com-
mittees that were both open to persuasion and had a high probability of
influencing the ultimate policy outcome. So, which committees were both
important and open to persuasion? Answering this question requires an
independent measure of importance or persuadability.
In studying elections, surveys and ratings such as the Cook Report and
Congressional Quarterly provide a measure of preelection competitiveness,
while, after the fact, election results show how close the race actually was.
Unfortunately, unlike election results, independent information on the im-
portance and persuadability of a committee and its members does not sim-
ply come off a library shelf. There are no surveys of the various consider-
ations bouncing around a legislator's mind, and a dichotomous yes or no
vote on the floor or in committee tells us little about how competitive an
individual legislator's decision actually was.
Therefore, in an effort to provide some independent yardsticks, I supply
four measures (admittedly imperfect) of which committees were competi-
tive and which committees were likely to convey the most useful information
to the full House and Senate. In spite of the fact that the committees all had
Democratic majorities, the proportions differed. Since Republican opposi-
tion to the Clinton plan was virtually unanimous, one should expect com-
mittees with less pronounced Democratic majorities to draw the attention of
interest groups. Moreover, although vote scores tell us little about attitudes
on specific issues and should be treated with caution, we should expect
committees with more moderate compositions to draw more attention. Al-
though not a guarantee of eventual support, cosponsorship of a bill certainly
indicates an initial level of support for a piece of legislation. Therefore,
committees with large numbers of cosponsors of the Clinton plan should
have drawn relatively less attention. Finally, we should also expect commit-
tees that could not afford to lose the votes of very many Democratic mem-
bers to have been the focus of attention. Table 5.3 summarizes information
on the percentage of Democratic members, mean Americans for Democra-
tic Action (ADA) scores for those members, the percentage of committee
members cosponsoring the Clinton plan, and the percentage of Democrats
needed to report out legislation.
Of the House committees, the Education and Labor Committee had the
largest proportion of Democrats, the greatest number of cosponsors of the
Clinton bill, and, according to ADA scores, the most liberal voting records. 3
Moreover, even with no Republican help, Chairman William Ford could
afford to lose more than one in five of his Democratic members and still
report legislation out of his committee. Conversely, the Ways and Means
Committee as well as the Energy and Commerce Committee had member-
ships that were less Democratic, more moderate, and their chairmen (Dan
Rostenkowski and John Dingell) could afford to lose fewer Democratic
3
ADA scores are one of the most commonly used ideological ratings. Other frequently used
ratings are those by the American Conservative Union and the National Journal. All the
measures correlated virtually perfectly.
Table 5.3. Measures of Committee Competitiveness
Mean ADA Score Committee Democrats

Democratic for Democratic Cosponsoring Clinton Democrats
Committee Members Members Plan Needed
House Energy and Commerce 27/44(61%) 77 10/27 (37%) 23/27 (85%)
House Ways and Means 24/38 (63%) 80 9/24 (38%) 20/24 (83%)
House Education and Labor 28/43 (65%) 85 15/28(54%) 22/28 (79%)
Senate Finance 11/20(55%) 76 3/20(15%) 11/11 (100%)
Senate Labor and Human Resources 10/17 (59%) 87 11/18(53%) 9/10(90%)
Source: Author's calculations from 1994 Almanac ofAmerican Politics (Barone and Ujifusa 1995).
Table 5.4. Committee Targets
Groups Mobilizing
Committee in Committee
House Energy and Commerce 20
House Education and Labor 4
House Ways and Means 15
Senate Labor and Human Resources 5
Senate Finance 16
votes. In other words, opposition interest groups needed to pick off fewer
Democrats to win in the latter two committees.
In the Senate, comparing the Finance Committee with the Labor and
Human Resources Committee yields a similar picture. Finance Committee
Chairman Daniel Patrick Moynihan could not afford a single Democratic
defection if he was to report out legislation. Furthermore, the Labor and
Human Resources Committee had more liberal Democrats and cosponsors
of the Clinton plan than did the Finance Committee. In fact, a majority of
Labor and Human Resources Committee members were cosponsors of the
Clinton plan.
In total, these independent measures, taken together with my theory of
mobilization choice, point to the Energy and Commerce Committee and the
Ways and Means Committee in the House and the Finance Committee in
the Senate, as the most likely targets.
The information gathered in my interviews is consistent with these expec-
tations. As Table 5.4 illustrates, of the groups with whom I spoke, attention
paid to the committees fell into two distinct categories. The House Energy
and Commerce Committee, the House Ways and Means Committee, and
the Senate Finance Committee received three to four times as much atten-
tion as the Senate Labor and Human Resources Committee and the House
Education and Labor Committee.
My interviews also permit us to look beyond these quantitative measures.
Both opponents and supporters of the Clinton plan gave me a similar expla-
nation for the pattern of committee targeting during the legislative battle
over health care reform. Liberal Democratic majorities on Ted Kennedy's
Senate Labor and Human Resources Committee and William Ford's Edu-
cation and Labor Committee insured that legislation mirroring the Clinton
plan would be reported out of these committees. Consequently, supporters
and opponents alike paid scant attention to the proceedings in those com-
mittees. Four of the six different groups that lobbied in the Education and
Labor and Labor and Human Resources Committees were to the left of
center. Their goals were to add more benefits or get a single-payer plan
passed in the two most liberal committees.
When I asked an insurance lobbyist why his company did not target the
Senate Labor and Human Resources Committee or the House Education
and Labor Committee, he quipped, "We weren't going to pay any attention
to those people's republic committees. There were too many coconspirators
(i.e., cosponsors) in those committees." Speaking just about the House,
Johnson and Broder put it in a little less colorful language:
There were only two places to pass a bill that could pass the House. Ways and Means
was one; Energy and Commerce was the other. The third committee with juris-
diction - Education and Labor - was a liberal bastion; over the years, the big unions
that were major bankrollers of Democratic Congressional candidates, including the
National Education Association and the American Federation of Teachers, had
packed that committee with people of unquestioned loyalty. "Ed and Labor" was
certain to report both the administration bill and the single payer plan favored by
many of the unions. But Ed and Labor was well to the left of the House. It had no
jurisdiction over financing provisions and only a marginal claim on Medicare and
Medicaid, the major existing health care programs. It was not likely to provide
legislation the leadership could take to the House floor. (1996, 306)
From these accounts, targeting decisions made by groups were consistent
with my expectation that lobbyists were likely to target those committees
that were likely to provide the best information to their respective bodies and
most likely to influence the ultimate legislative outcome. My interviews
provided evidence that this really was the case. Interest groups targeted
committees where the outcome was in doubt and where their efforts could
make a difference.
Although the variation in committee targets is certainly noteworthy, the
general targeting of committees may not seem to be a very surprising find-
ing. After all, as discussed in Appendix C, most of the action regarding the
Clinton plan occurred in committee, and health care legislation was never
even brought to the floor of either house for a vote. But when we remember
some additional particulars of the rules and procedures in the respective
bodies, the fact that there was any attention paid at all to committees pro-
vides evidence for an information model of lobbying and for my theory of
mobilization choices.
Specifically, Senate rules allow only one committee to have primary juris-
diction over a particular piece of legislation, and neither Kennedy's Labor
and Human Resources Committee nor Moynihan's Finance Committee
was willing to cede responsibility. According to a number of accounts,
George Mitchell planned to use his power under Senate rules as majority
leader to bring health care legislation directly to the floor. On the House
side, the leadership could work through the Rules Committee to reshape any
bill that was voted out of any committee {Congressional Quarterly Almanac
1994, 320). The Education and Labor Committee was virtually guaranteed
to report out a bill that could be shaped by Majority Leader Gephardt and
Speaker Foley in the House. Why then bother to mobilize constituent com-
munications during the committee stage?
The explanation is that committees provide information and are not sim-
ply procedural hurdles that need to be surmounted. In addition to passing
and shaping legislation, the mandate of a committee is to provide political
and policy cues (Krehbiel 1991). Committees work out both the politics and
the policy of a bill and provide key information to each respective body as a
whole. Accordingly, a committee's importance can be judged by the quality
of information it conveys. In the case of health care, my theory predicts that
interest groups should have targeted committees where they had a chance of
influencing the ultimate committee outcome as well as the ultimate legisla-
tive outcome. The House Education and Labor Committee and the Senate
Labor and Human Resources Committee were poor targets because their
memberships were not open to persuasion and their decisions were not
likely to carry much weight in the Congress as a whole. Consequently,
health care bills recommended by committees with liberal majorities would
have provided little information to colleagues and the congressional leader-
ship crafting a floor version. As an aide to Majority Leader Mitchell told me,
"Kennedy's committee was simply not going to give us enough useful intelli-
gence about the politics of health care."
Legislator Targets
Within the key committees not every member was targeted. As is expected,
with a grass roots tactic having a legislative objective, members considered
"undecided" or "persuadable" were most likely to be the targets of grass
roots lobbying efforts.4 Table 5.5 reports the individual targets of groups.
Three groups in my sample targeted supporters who needed help explain-
ing their opposition to the Clinton plan to the administration and to the
House Democratic leadership. Two groups took advantage of the political
environment surrounding the health care situation to build membership.
The large majority of groups, however, targeted their grass roots efforts
solely at members they considered to be undecided. "Hunt where the ducks
4
Many of the groups I studied utilized more nuanced measures of legislators' predispositions.
For instance, many used a one-through-five scale. In cases where I was given more detailed
information, I collapsed thefive-pointscale into a three-point scale, counting soft supporters
and opponents as undecided.
Table 5.5. Legislative Targets within Committee
Targets Number of Groups

Supporters 3 (21%)
Opponents 2 (17%)
Undecideds 16 (67%)
are," was how one lobbyist explained his tactics to me. He added: "It was
good old-fashioned vote counting. We worked with our Hill lobbyists and
other allies to determine who was with us and who was against us." Another
who was lobbying for the Clinton plan remarked, "We figured out who was
for us and who was against us and threw the kitchen sink at everyone in the
middle." Pam Bailey of the Healthcare Leadership Council targeted her
group efforts to "Specific committees and specific votes" and to "Tradi-
tional swing districts" (Stone 1994).
West, Heith, and Goodwin reported the following observations by Bill
Gradison, executive director of the HIAA:
We moved our ads around focused on the districts where key committee members
were and also on key geographical areas. We felt the battleground would be the
Border States and the Southern ones. That's where there were the most conservative
Democratic members who we thought would be sympathetic to our message. . . .
We moved around our focus depending on what committee or subcommittee was
considering a bill at a particular time. In that sense, it was highly focused. We tried to
reach the swing members of those committees on a serial basis depending on when
they would be taking things up. (1995, 11, 24)
The nature of my sample, which surely overrepresented large resource-

rich groups, clearly affected the influence of organizational factors on lobby-
ing choices. A sample reflecting more accurately all the groups that mobi-
lized around health care would have provided more analytical leverage on
the topic of organizational resources and mobilization choices. Still, some
speculative assessments on the influence of organizational resources and
organizational characteristics on where groups decided to target can be
drawn.
On the one hand, trade associations, insurance companies, pharmaceuti-
cal companies, and unions all had widespread membership and potential
allies virtually everywhere - as well as the money to activate them. Even so,
organizational constraints and lack of resources did hamper some groups.
For example, a major coalition of pro-reform groups did not have adequate
funds and was forced to target many districts with no short-term legislative
payoff in an effort to raise money. As an organizer from one of these groups
put it: "We were too busy recruiting and trying to keep our heads above
water to target effectively. For instance, we had a thousand watch parties
when Clinton gave his speech where we had people write their congressman.
We tried to spin that in the media as a sign of strength. But it really was a sign
of weakness. Most of these people were writing to members who were
already with us."
In general, lobbyists trying to mobilize support for health care reform had
a tough time. This was the case even for groups with sufficient resources.
For example, the American Association of Retired Persons (AARP) and
unions were never really able to mobilize the type of support that the admin-
istration had expected. According to one administration official, the AARP
thought that reform was a done deal and spent too much time trying to fine-
tune it. As for the unions, the leadership was supporting the Clinton plan,
but the rank and file were skeptical. Most already had good health care plans
and had previously sacrificed salary increases to get them. Unions ended up
pursuing less focused and less effective campaigns because their member-
ship was not solidly behind them. In addition, unions were left "organiza-
tionally exhausted" by their fight against NAFTA.5
Another group, the Christian Coalition, used the battle over health care as
a recruiting device. Handing out hundreds of thousands of postcards in
churches all over the country, it mobilized in many districts whose members
already had taken strong positions on the Clinton plan. The postcards they
distributed had a Norman Rockwell painting of a mother, child, and doctor.
In big letters was the exhortation, "Don't let a government bureaucrat in on
this picture." The card then folded out with two preaddressed cards to the
worshiper's senators and one to his or her representative. The most impor-
tant card, though, was the one to be sent back to Christian Coalition head-
quarters. This card contained the worshiper's name, address, and phone
number. This information could then be used for future fund-raising or
mobilization campaigns around other issues.
Still, notwithstanding the organizational hurdles that some groups faced
and the organization building that some interest groups pursued, the data
from interviews with interest groups involved in health care grass roots
lobbying are consistent with the theoretical expectations of my model. In-
terest groups targeted undecided members situated in key legislative posi-
tions, where their grass roots lobbying efforts were most likely to make a
difference.
5
See Skocpol (1996) for a summary of the major players and their actions during the debate
over the Clinton plan.
86 Interest Groupsy Lobbying, and Participation
Constituent Targets
Targeting efforts to stimulate constituent communications is a two-stage

process. Once particular districts or states are chosen, particular constitu-
ents within those districts need to be identified and mobilized. With a legis-
lative objective, my theory predicts that those with the highest probability of
influencing a member of Congress should be the constituents most likely to
be targeted. Often, these are constituents with greater resources and more
personal connections to a particular member. Yet, they can also be those
constituents who possess the credibility to convey particular types of infor-
mation. Using the data gathered in my interviews, Table 5.6 contains the
reports of interest groups on their constituent targets.
Consistent with my expectations about mobilization choices, groups were
likely to target those over whom they had leverage (employees and sup-
pliers). For example, IBM, Eastman Kodak, Mobil, DuPont, and Xerox
conducted large-scale campaigns to stimulate communications to Congress
in targeted districts (Mintz 1994). In-house letters, e-mails, and voice mail
messages were some of the tools these companies utilized to contact their
employees. Middle management and supervisors were then responsible for
follow-up and insuring that contacts were made. One major insurance com-
pany targeted all its suppliers and made it known in a not-so-subtle way that
each supplier's response would be factored into future purchasing decisions
by the company.
Groups also targeted reliable supporters in the form of members and
stockholders. Concerned about price caps on prescription drugs, phar-
maceutical companies with whom I spoke were especially aggressive about
"educating" and targeting stockholders on health care issues. One major
pharmaceutical company matched its lists of stockholders and retirees (of-
ten large stockholders) to Federal Elections Commission contributor lists
and then matched these lists to their correct political geography to deter-
mine the "assets" they possessed to influence targeted members of Con-
gress. A group of pharmaceutical companies also conducted focus groups
and surveys of stockholders to see which particular messages were most
likely to get people to contact their member of Congress about health care
legislation.6 Using their enhanced lists and the messages developed from
extensive research, the companies then hired a consulting firm to contact
these targeted stockholders. In all, nineteen thousand retired workers and
6
Such research was obviously not unique to the pharmaceutical industry. Virtually every
component of the Clinton health care plan was researched with surveys and focus groups.
Furthermore, research to gauge the effectiveness of various messages was not only valuable
in mobilizing particular constituents but in refraining the issue in general.
Table 5.6. Constituent Mobilization Targets and Levels of

Leverage
Mobilization Targets Number of Groups

Members/Stockholders 13
Employees/Vendors 7
Third Party 8
Mixed 7
stockholders were patched through to twenty different Congressional

offices.7
Contacting and recruiting those with a greater probability of influencing a
member of Congress were the priorities of many groups. As a government
relations official with a pharmaceutical company put it, "When a member of
Congress goes home, they spend ninety percent of their time with one
percent of the population. We wanted that one percent to be people who
were going to communicate our message." The NFIB put in place a pro-
gram called Guardian Advisory Council (GAC). A GAC team in a targeted
congressional district comprised between five and ten influential business
people. Along with the more broad-based efforts to mobilize all NFIB mem-
bers (already an influential group), the GAC teams - composed of larger
employers as well as the college friends, neighbors, and former colleagues of
the targeted legislator - were also deployed.
Although key contacts were certainly emphasized, there was also a signifi-
cant level of more broad-based or third-party lobbying campaigns. This is
evidenced in Table 5.7, which reports the constituent targets of interest
groups.
Yet even these campaigns (in which groups used television or radio adver-
tising to stimulate communications to Congress from constituents outside
of their family of members, employers, suppliers, and stockholders) were
not shotgun efforts. Particular programs and stations were chosen to maxi-
mize the probability that the mobilization requests would not fall on deaf
ears, that the correct message would be conveyed, and that the communica-
tion would matter. Ben Goddard, who produced and placed the famous
7
Under current lobbying regulations, the company was not required to disclose any of these
efforts.
Table 5.7. Constituent Mobilization Targets
Mobilization Targets Number of Groups

Broad Based 15
Key Contacts 11
Mixed 5
"Harry and Louise" advertisements for the HIAA, explained, "Our media
buys were targeted on involved Americans, people who were registered to
vote, wrote letters to editors or public officials, attended meetings and made
political contributions.8 We bought time on CNN and Headline News,
CNBC, and Rush Limbaugh" (West et al. 1995, 10).
The Health Insurance Association of America also went for a broader-
based strategy because it believed that it needed more credible messengers. 9
An HIAA staff member explained to me, "We were the black hats of the
health care game. People like doctors, nurses, and small business. They
don't like insurance companies. We needed to work through others to get
our message listened to." In its targeted districts the HIAA hired field direc-
tors. "These were not just college kids. Our field people had extensive
political and public relations experience. Their job was to cultivate allies,
gather lists, and find those who would be on our side." The HIAA used a
Washington-based consultant to follow through on the responses stimulated
by "Harry and Louise" - patch-thrus and telegrams - and relied on their
field operatives to stimulate communications from the grass tops.
Both Marty Plisner of CBS News and David Broder of the Washington Post
pointed out to me that the Republican and antireform groups enjoyed quite
an advantage in hiring field operatives. Because the Republicans were out of
the White House, there was a real talent pool of Republican campaign and
public relations operatives and a lot of highly skilled political people were
looking for jobs. For example, the NFIB hired Marc Nuttle, who was the
field director for Pat Robertson in his 1988 race for president and is one of
the most highly regarded field organizers in GOP circles.10 On the Demo-
8
Goddard recently married the actress who played Louise.
9
The actual HIAA lobbying plan was leaked and was widely available in Washington, D.C. It
was the subject of numerous media reports, including a lengthy report by Brooks Jackson on
CNN (October 19, 1993).
10
See Kosterlitz (1993) and Weisskopf (1993) for accounts of the scramble for political talent
and field operatives in the months preceding the unveiling of the Clinton plan.
cratic side, many of the people with strong field and grass roots experience
were enjoying administration jobs.
In short, judgments about the influence, credibility, and direction of a
communication appear to have influenced targeting decisions at the individ-
ual level. Organizational resources and characteristics, however, also appear
to have had a significant influence. Groups made efforts to identify reliable
supporters who could influence the legislative process. If they did not have
such supporters or if the communications of their reliable supporters would
not have been credible - and if they had the monetary resources to do so -
they pursued broader-based tactics.
Summary
This overview of mobilization choices made during the legislative battle over
health care reform provides confirmatory evidence that mobilization choices
were largely a function of group estimates of three characteristics: the per-
suadability of a legislator, the importance of a legislator, and the influence of
citizens. Organizational constraints seem to have influenced the decisions of
some lobbyists on the pro-reform side, but the nature of my data makes it
difficult to draw any definitive conclusions.
Although the preceding discussion provides a good description of the
basic lobbying decisions, my interviews and observations permit me to do
more than simply provide descriptive statistics and a basic analysis of tactical
choices. Because I gathered more detailed information on the exact targets
and messages of groups, I am able to examine in much greater detail the
legislators and constituents who were targeted in order to influence the work
of three key committees: House Energy and Commerce, House Ways and
Means, and Senate Finance.
Energy and Commerce

During the legislative debate over the Clinton plan and health care reform,
John Dingell of Michigan chaired the Energy and Commerce Committee.
Elected to Congress in 1955 at the age of twenty-nine to succeed his father,
Dingell was one of the most senior members of the U.S. House of Representa-
tives. He was a longtime champion of national health care insurance. In fact,
at the start of every congressional session, he had made it a habit to submit
legislation calling for health coverage for all Americans. When the Democrats
controlled Congress, he was considered one of the most powerful committee
barons and his committee one of the most powerful. Energy and Commerce
handled four out of every ten bills in the House. It had the largest budget and
staff of any committee in the House (Barone and Ujifusa 1995).
Furthermore, the Energy and Commerce Committee was considered by

lawmakers and lobbyists alike to be representative of the body as a whole
and a key cue giver to other members of Congress. "If Commerce couldn't
figure this out, nobody could. It was the bellwether committee," explained
one business lobbyist. Ira Magaziner, chief of the Clinton health care task
force, remarked, "All our political people were telling us that Energy and
Commerce was the most important committee - that it most reflected the
membership of the House."
In the 103rd Congress, the Energy and Commerce Committee was com-
prised of twenty-seven Democrats and seventeen Republicans. On health
care legislation, initial head counts indicated that Dingell could count on no
support at all from Republicans. Thus, he needed the support of twenty-
three out of twenty-seven of his Democratic colleagues. My interviews on
the Hill, confirmed by contemporaneous media reports, suggested that of
those twenty-seven Democrats, Dingell believed he could count on seven-
teen {Congressional Quarterly Almanac 1994). The Congressional Quarterly
Almanac handicapped the ten remaining Democrats as follows (1994, 335):
four members were leaning strongly against the Clinton plan (Ralph Hall of
Texas, Billy Tauzin of Louisiana, Roy Rowland of Georgia, and rival-plan
author Jim Cooper of Tennessee), three were leaning toward the plan (Lynn
Schenk of California, Marjorie Margolies-Mezvinsky of Pennsylvania, and
Blanche Lambert of Arkansas), and three were up-for-grabs (Jim Slattery of
Kansas, Rick Boucher of Virginia, and Richard Lehman of California).
In Table 5.8, I report three other measures of a member's likelihood of
being targeted for a mobilization campaign: the NFIB's handicapping of the
Energy and Commerce Committee in the fall of 1993, whether or not a
member was a cosponsor of the Clinton plan, and ADA scores. The NFIB
system is identical to one that would be used to identify competitive elec-
tions or voters in a get-out-the-vote effort: 1 denotes a strong supporter of
the Clinton plan; a 2, a member leaning toward the Clinton plan; a 3, a pure
undecided; a 4, a member leaning against the Clinton plan; and 5, a strong
opponent.
None of these measures is a perfect indicator of a member's predisposi-
tions. Cosponsors have been known to change their minds and general
ideological scores may say little about attitudes on particular issues. More-
over, the NFIB measure is the assessment of only one group and deals
specifically with legislators' predispositions toward the employer mandate in
the president's plan. Nevertheless, all else held equal, cosponsors are more
likely to support a bill; and ideological scores provide a basic assessment of
members' predispositions. In addition, the employer mandate was the main
bone of contention in the committee, and the NFIB rankings are virtually
identical to targeting assessments that I examined from other groups. In
sum, although no individual measure is perfect, to the extent that each
indicator points in the same direction, we can be confident in the entire set's
utility as a measure of swing legislators.
Given the NFIB estimates, my model of grass roots lobbying decisions
predicts that seven Democratic members of the Energy and Commerce
Committee should have received the most attention: Representatives
Boucher, Cooper, Margolies-Mezvinsky, Lambert, Lehman, Schenk, and
Slattery. Moderate members (according to the ADA ratings) Boucher,
Cooper, Klug, Lambert, Slattery, Tauzin, and Upton can be judged as likely
targets. In addition, we should expect the ten cosponsors of the Clinton plan
to have received little or no attention from grass roots lobbyists. (Signifi-
cantly, none of the previously mentioned legislators was cosponsor.) If we
look at a measure of member vulnerability (and thus a need for information
on electoral consequences), Representatives Boucher, Cooper, Lambert,
Lehman, Margolies-Mezvinsky, Pallone, and Schenk all represented dis-
tricts that voted for Bush in 1988 and that were barely carried by Clinton in
1992. Rowland and Hall represented districts that were carried by Bush in
both 1988 and 1992.
To test how these expectations stacked up against the actual targeting
decisions of lobbyists, I asked the representative from each interest group in
my study to name the five members on whom they expended the most grass
roots lobbying efforts. Of the twenty-one groups with whom I spoke, twenty
targeted members of the Energy and Commerce Committee and fifteen
were willing or able to give me the breakdown on their mobilization efforts.
Their behavior is reported in Table 5.9.
The interview data confirm that swing members of the Energy and Com-
merce Committee were in fact the ones most likely to be targeted for grass
roots mobilization efforts. These data are also consistent with the stylized
story that one hears about the fight in the Energy and Commerce Commit-
tee. According to the public and private accounts of various interest groups,
congressional staffers, White House officials, and members of the press,
Chairman Dingell, by April, had managed to convince three of the six
undecideds: Representatives Margolies-Mezvinsky, Schenk, and Lambert.
At this point, he had twenty votes in hand. Needing three more votes, his
attention turned to Representatives Boucher, Lehman, and Slattery. Dingell
knew that he needed the support of all three and, as my data demonstrate, so
did interest groups.
The tone of the messages used to stimulate constituent communications
very much concentrated on short-term direct costs and fears. The following
examples illustrate some of the mobilization efforts and advertisements that
Table 5.8. Handicapping House Energy and Commerce Committee
Member NFIB Ratings CO! ADA Ratings

Joe Barton, TX-R 5 No 0
Michael Bilirakis, FL-R 5 No 20
Thomas J. Bliley, VA-R 5 No 5
Rick Boucher, VA-D 3 No 60
Sherrod Brown, OH-D 1 Yes 95
John Bryant, TX-D 1 No 90
Cardiss Collins, IL-D 1 Yes 100
Jim Cooper, TN-D 4 No 55
Michael Crapo, ID-R 5 No 5
John D. Dingell, MI-D 1 Yes 70
Jack M. Fields, TX-R 5 No 0
Gary A. Franks, CT-R 5 No 10
Paul E. Gillmor, OH-R 5 No 15
Jim Greenwood, PA-R 5 No 20
Ralph M. Hall, TX-D 4 No 20
Dennis Hastert, IL-R 5 No 5
Scott Klug, WI-R 4 No 40
Mike Kreidler, WA-D 1 Yes 90
Blanche M. Lambert, AR-D 2 No 65
Richard H. Lehman, CA-D 3 No 75
Thomas J. Manton, NY-D 1 No 70
Edward J. Markey, MA-D 1 Yes 95
Alex McMillan, NC-R 5 No 10
Marjorie Margolies-Mezvinsky, PA-D 2 No 80
Carlos J. Moorehead, CA-R 5 No 0
Michael G. Oxley, OH-R 5 No 5
Frank Pallone Jr., NJ-D 2 No 90
Bill Paxon, NY-R 5 No 5
Bill Richardson, NM-D 1 Yes 75
Roy Rowland, GA-D 4 No 20
Dan Schaefer, CO-R 3 No 5
Lynn Schenk, CA-D 2 No 85
Philip R. Sharp, IN-D 1 No 85
Jim Slattery, KS-D 3 No 60
Clifford B. Stearns, FL-R 5 No 15
Gerry E. Studds, MA-D 1 Yes 90
Al Swift, WA-D 1 Yes 95
Michael L. Synar, OK-D 1 Yes 85
J. Tauzin, LA-D 4 No 35
Edolphus Towns, NY-D 1 No 95
Fred Upton, MI-R 4 No 35
Craig Washington, TX-D 1 No 90
Henry A. Waxman, CA-D 1 Yes 100
Ron Wyden, OR-D 1 No 95
Sources: NFIB ratings from author's interviews, ADA ratings from Barone and Ujifusa (1993).
Table 5.9. Member Targets in House Energy and Commerce Committee
Member Groups Mobilizing

Jim Slattery, KS-D 13
Jim Cooper, TN-D 10
Rick Boucher, VA-D 9
Richard H. Lehman, CA-D 9
Lynn Schenk, CA-D 7
Marjorie Margolies-Mezvinsky, PA-D 7
Blanche M. Lambert, AR-D 6
J. Tauzin, CA-D 4
Scott Klug, WI-R 3
Fred Upton, MI-R 2
Ralph M. Hall, TX-D 1
Total 15
were used by interest groups to stimulate communications against the Clin-

ton plan in the districts of swing members of the Energy and Commerce
Committee.
Citizens for a Sound Economy aired the following radio spot in the
districts of Boucher, Lehman, and Slattery.
Announcer: Medical care in America under health care reform [as telephone rings in
background].
Man: Gatekeepeer.
Woman: Hello, this is Mrs. Baylor. My son is having a terrible earache and needs to
see Dr. Murray right away.
Man: You will not see Dr. Murray. Dr. Johnson will see your son next week.
Woman: Next week? He needs a doctor now! Is Dr. Johnson an ear specialist?
Man: It doesn't matter.
Woman: Wait a minute. I have health insurance. I don't need you.
Man: Mrs. Baylor, under health reform all Americans - and that includes you and
your son - will have to go through government health alliances with gatekeepers
like me. We will decide who, when, or even if you see a doctor.
Woman: But this is America!
Announcer: If you don't want government gatekeepers telling you what doctor you
can see, call Congressman [name] at [number] and tell him to vote no on the
Clinton health care plan. That's [number].
The next example comes from a piece of direct mail aimed at getting small
business owners in Richard Lehman's district to contact him. It was sent by
the NFIB.
BACK ROOM HEALTH CARE DEAL CUT!
The committee's chairman is working furiously behind closed doors to cut a deal to
get Rep. Lehman to support an employer mandate that could be bad for your small
business. Call him now at (202) 225-1000 and tell him that the employer mandate
will kill jobs and your business.
Interestingly, pro-reform groups were frustrated by the fact that the mes-
sage that had tested best for them was an anti-Congress one - " y ° u deserve
the same health care benefits as Congressman X" - that was killed by a
skittish House Democratic leadership. This topic was the subject of much
debate among political consultants on the Democratic side. The "every
American deserves health care as good as Congress gets" message was
eventually used very late in the battle.
Particularly strong attention was paid to Jim Slattery of Kansas. Slattery
was running for governor of Kansas and, according to many accounts, was
the deciding vote on employer mandates. An action alert sent out by the
NFIB by fax and e-mail made the following plea:
Representative Jim Slattery could be the deciding vote on the powerful House En-
ergy and Commerce Committee on the proposal to require you and all small busi-
ness owners to pay for a government mandated health benefits package for all
employees - full time, part time, and seasonal. I need your help to convince him that
an employer mandate of any type would be devastating to Kansas small business and
your employees. Please call or write to Congressman Jim Slattery today before it's
too late! Please do not send a copy of this action alert to Rep. Slattery. Your own
letter is much more effective.
Finally, Johnson and Broder (1996) reported the following series of
memos from NFIB's House lobbyist Mark Isakowitz to the group's presi-
dent Jack Faris. They provide an excellent illustration of the sort of tactical
logic I have been discussing.
February 12: Rep. Jim Slattery who could be the tie-breaking vote in the Health
Subcommittee on the mandate met with Hillary Clinton. Prior to the meeting his
staff had told us he had decided to oppose the mandate. After the Hillary meeting,
his staff said he was still open to compromise on the mandate. Our latest anti-
mandate Action Alert just went out in his district.
March 4: Kim heard from a good source that Rep. Jim Slattery of KS would offer a
compromise 50 percent employer mandate in the Commerce Committee. Eight
GAC members are going to see him on Monday.
March 11: Eight Kansas GAC members met with Rep. Jim Slattery in Topeka on
Monday. He is a key swing vote on Commerce and he told our members he
believes there should be some employer mandate in the bill. He is running for
governor so there might be some additional grass roots we can do there.
March 25: Rep. Jim Slattery appears to be the major deal maker with Chairman
Dingell on this compromise plan. Today, an Action Alert to ALL Kansas NFIB
members is going out specifically on Slattery. We are going statewide with the alert
because he is running for Governor and has to be sensitive to concerns outside his
district. Kim and I met with Slattery on Tuesday.
April 8: The day before Clinton arrived in Topeka to campaign for health care reform
with Rep. Slattery, a group of business owners (including an NFIB member)
unveiled the ad to appear in the next day's Topeka paper announcing the coalition
of more than 1 million employers opposing the mandate.
April 15: Chairman John Dingell continues to negotiate with the people we have been
targeting but does not yet have a majority on the Commerce Committee. There
are signs our grassroots are working. Slattery has retreated some on the mandate,
and is said to have proposed this week a way to get universal coverage without the
mandate; he told our members in March this was impossible.
April 22: Rep. Jim Slattery announced his opposition to the employer mandate
on Thursday saying he was against it and that he wouldn't vote for it even to just
move the process forward in the Commerce Committee. As you know, this is a
complete reversal. We got some credit for it in The Washington Post today. (1996,
341-342)
Ultimately, Dingell was not able to convince enough of his Democratic

colleagues. Health care reform never even came to a vote in the Energy and
Commerce Committee. Magaziner lamented, "The NFIB really killed us in
Energy and Commerce. When they won in Dingell's committee, that really
gave heart to our opponents."
Assessments of causality and the granting of credit and blame by the
Washington Post and Magaziner may or may not be accurate. Still, the mobili-
zation decisions made in the Energy and Commerce Committee during the
legislative battle over the Clinton plan and health care reform are an excel-
lent illustration of the careful targeting attention paid to swing members of
key committees. The preceding discussion demonstrated that if we are to
evaluate lobbying influence we need to understand mobilization strategies
and tactics, and the specific political environment in which these decisions
are made. Furthermore, if we are to understand mass participation - if we
are going to understand who participated in Kansas during the battle over
health care reform - we must pay attention to more than personal resources,
attitudes, and attachments.
House Ways and Means Committee

Dan Rostenkowski's Ways and Means Committee was another battleground
in the fight over health care. Although there were many thorny issues, the
central debate revolved around the employer mandate. With Republicans
on the committee unanimously opposed to the employer mandate, Ros-
tenkowski needed the support of twenty out of the twenty-four Democrats
on his committee. Consequently, Rostenkowski worked hard to keep his
members in line and to shield them from constituent pressure. In fact,
after the HIAA's airing of their "Harry and Louise" commercials during
the 1994 February recess unleashed a flood of constituent communica-
tions in the districts of some of his key members, Rostenkowski reached
an accord with the HIAA to stop agitating the constituents of members
of the Ways and Means Committee (Johnson and Broder 1996; Scarlett
1994).
By the time the bill reached markup in the full committee, however,
Rostenkowski had been indicted on a series of federal fraud charges and was
no longer chairman. The task of shepherding health care reform through
the Ways and Means Committee fell to his successor, Representative
Sam Gibbons of Florida. There was already a significant amount of at-
tention being paid to the Ways and Means Committee. Nevertheless, ac-
cording to a number of Hill staffers with whom I spoke, the fall of
Rostenkowski made Ways and Means an even more attractive target. Mem-
bers simply did not owe or fear Chairman Gibbons as much as they had
Chairman Rostenkowski. Moreover, with Rostenkowski's demise and Gib-
bons's ascension, the deals with HIAA and other interest groups broke
down, and the "Harry and Louise" commercials started up once again in
the districts of swing members of the committee (Johnson and Broder
1996).
The playing field that interest group lobbyists faced is portrayed by the
information in Table 5.10. Along with the NFIB ratings, cosponsor status,
and ADA scores, I also include whether a member ultimately supported the
bill to come out of committee. To judge from the NFIB scores, Michael
Andrews of Texas, Bill Brewster of Oklahoma, Peter Hoagland of Nebraska,
and L. F. Payne of Virginia should have been prime targets. Looking at ADA
scores, Andrews, Brewster, Hoagland, Payne, J. J. Pickle of Texas, and
Nancy Johnson of Connecticut should have been the focus of mobilization
efforts. The eight cosponsors of the Clinton plan should not be expected to
have received much attention. (None of the names previously mentioned as
targets were cosponsors.) Andrews can be excluded from the list because
information on the electoral consequences of health care reform was of no
use to him. He had lost a March race for the Democratic Senate nomination
in Texas and was not running for reelection to the House. Similarly, Pickle
was retiring and was not likely to be susceptible to mobilization pressure or
in need of the information that it provided. In sum, three names come up in
all the measures: Brewster, Hoagland, and Payne. Hoagland and Payne
represented districts won by Bush in both 1988 and 1992. Brewster's
district was won by Bush in 1988 and only narrowly carried by Clinton in
1992.
Fifteen groups with whom I spoke targeted the Ways and Means Commit-
tee and thirteen of these groups were willing and able to provide me with
Table 5.10. Handicapping House Ways and Means Committee
Member NFIB Ratings Cosponsor ADA Ratings Actual Vote

Michael A. Andrews, TX-D 3 No 55 No
Bill Archer, TX-R 5 No 5 No
Bill Brewster, OK-D 3 No 25 No
Jim Bunning, KY-R 5 No 10 No
Dave Camp, MI-R 5 No 5 No
Benjamin L. Cardin, MD-D 1 Yes 90 Yes
William J. Coyne, PA-D 5 Yes 95 No
Philip M. Crane, IL-R 5 No 15 No
Harold E. Ford, TN-D 1 No 80 Yes
Sam M. Gibbons, FL-D 1 No 75 Yes
Fred Grandy, IA-R 4 No 20 No
Mel Hancock, MO-R 5 No 5 No
Wally Herger, CA-R 5 No 5 No
Peter Hoagland, NE-D 3 No 65 No
Amo Houghton, NY-R 4 No 20 No
Andy Jacobs Jr., IN-D 1 No 90 Yes
William J. Jefferson, CA-D 2 No 95 Yes
Nancy L. Johnson, CT-R 4 No 40 No
Barbara B. Kennelly, CT-D 2 No 90 Yes
Gerald D. Kleczka, WI-D 1 No 80 Yes
Mike Kopetski, OR-D 1 No 85 Yes
Sander M. Levin, MI-D 1 Yes 95 Yes
John Lewis, GA-D 5 No 100 Yes
Robert T. Matsui, CA-D 1 Yes 80 Yes
Jim McCrery, LA-R 5 No 0 No
Jim McDermott, WA-D 1 No 85 No
Michael R. McNulty, NY-D 2 No 60 Yes
Richard E. Neal, MA-D 2 No 95 Yes
L. F. Payne, VA-D 3 No 30 Yes
J. J. Pickle, TX-D 1 No 65 Yes
Charles B. Rangel, NY-D 1 Yes 95 Yes
Mel Reynolds, IL-D 1 Yes 90 Yes
Dan Rostenkowski, IL-D 1 Yes 90 Yes
Rick Santorum, PA-R 5 No 20 No
E. Clay Shaw, Jr., FL-R 5 No 10 No
Fortney H. Stark, CA-D 1 Yes 100 Yes
Don Sundquist, TN-R 5 No 5 No
William M. Thomas, CA-R 5 No 5 No
Sources: NFIG ratings from author's interviews, ADA ratings from Barone and Ujifusa (1993). Cosponsor and vote in
committee come from Legislate on-line service.
Table 5.11. Member Targets in House Ways and Means Committee
Member Groups Mobilizing

Peter Hoagland, NE-D 12
Bill Brewster, OK-D 10
L. F. Payne, VA-D 10
William J. Jefferson, CA-D 6
Richard E. Neal, MA-D 3
Barbara B. Kennelly, CT-D 3
Nancy L. Johnson, CT-R 2
Sander M. Levin, MI-D 2
Michael A. Andrews, TX-D 2
Total 13
targeting information. Table 5.11 shows the individual targets of groups

using mobilization tactics in the Ways and Means Committee.
Again, the findings are consistent with theoretical expectations. Hoag-
land, Brewster, and Payne - who by multiple indicators were the swing votes
on Ways and Means and the most in need of information on the political
consequences of their actions - were the members most likely to be targeted
by grass roots lobbyists. They were on the receiving end of advertisements
like the ones I outlined in the previous section on the Energy and Commerce
Committee. William Jefferson of Louisiana was also the focus of significant
lobbying attention. Although African American and generally considered a
liberal member of Congress, he was not a cosponsor of the Clinton plan and
had aspirations to run in a statewide race for governor. He was also an active
member of the Democratic Leadership Council. Connecticut members
Barbara Kennelly and Nancy Johnson (the only Republican on the list) as
well as Richard Neal of Massachusetts were targeted by home district insur-
ance companies.
The Ways and Means Committee eventually reported out a bill by
a two-vote margin (20 to 18). No Republicans voted in favor of the
legislation. Of the three members who received the most grass roots
lobbying attention, Hoagland and Brewster voted "no" and Payne voted
"yes." Jim McDermott, a liberal Democrat from Washington state and a
strong proponent of the single-payer option, voted against the Ways and
Means version because he felt it did not go far enough toward providing
universal health care coverage for all Americans (Clymer, Pear, and Toner
1994).
The Senate Finance Committee and the Mainstream Coalition

The Finance Committee in the Senate was perceived in very much the same
way as the Energy and Commerce Committee was in the House. The Con-
gressional Quarterly Almanac commented, "The Senate Finance Committee
occupied a pivotal place in the health care debate. Many lawmakers, espe-
cially in the House, were counting on the panel to craft a grand compromise
that could attract votes from both parties and pave the way for skittish
members to vote for an overhaul bill" (1994, 338).
Daniel Patrick Moynihan of New York chaired the Finance Committee.
Faced with a slim eleven to nine Democratic advantage in his committee,
Moynihan could not afford to lose even a single Democrat if he was to report
out legislation. Moreover, other accounts indicated that Moynihan was op-
posed to the strategy of just getting the legislation through with Democratic
votes. He knew that the real goal was sixty votes in the Senate and wanted to
craft a bill that would attract enough Republican votes to avert a filibuster.
Furthermore, although Democrats enjoyed only a slim majority in the com-
mittee, the atmosphere was less partisan than in the House and there were
moderate Republicans with whom Moynihan hoped to work.
Table 5.12 depicts measures of members' predispositions in the Senate
Finance Committee.
According to NFIB estimates, Senators Max Baucus of Montana, David
Boren of Oklahoma, John Breaux of Louisiana, and Kent Conrad of North
Dakota should have drawn the most attention of grass roots lobbyists. Sena-
tors Bill Bradley of New Jersey, John Danforth of Missouri, Dave Duren-
berger of Minnesota, and John Chafee of Rhode Island were also potential
targets according to the NFIB. ADA scores point in the direction of Sena-
tors Breaux (who was also very active in the Democratic Leadership Coun-
cil), Chafee, Danforth, Durenberger, Bob Packwood of Oregon, and Bill
Roth of Delaware. Although both Durenberger's and Danforth's ADA
scores and NFIB projections suggested that they could have been likely
targets, neither was running for reelection and therefore did not need infor-
mation on the electoral consequences of their actions. Because of Pack-
wood's problems with the ethics committee and his reliance on the Republi-
can leadership for protection, conventional wisdom was that Minority
Leader Bob Dole essentially controlled his vote. Although Baucus and Con-
rad were cosponsors of the Clinton plan, both had publicly expressed strong
reservations about the employer mandate (Johnson and Broder 1996).
Table 5.12. Handicapping Senate Finance Committee
Member NFIB Ratings Cosponsor ADA Ratings Actual Vote

Max Baucus, MI-D 3 Yes 85 Yes
David Lyle Boren, OK-D 3 No 70 Yes
Bill Bradley, NJ-D 2 No 90 Yes
John B. Breaux, LA-D 3 No 40 Yes
John H. Chafee, RI-R 4 No 55 Yes
Kent Conrad, ND-D 3 Yes 80 No
John C. Danforth, MO-R 4 No 35 Yes
Thomas A. Daschle, SD-D 1 Yes 35 Yes
Robert Dole, KS-R 5 No 10 No
Dave Durenberger, MN-R 4 No 75 No
Charles E. Grassley, IA-R 5 No 20 No
Orrin G. Hatch, UT-R 5 No 5 No
George J. Mitchell, ME-D 1 Yes 85 Yes
Daniel Patrick Moynihan, NY-D 1 Yes 90 Yes
Bob Packwood, OR-R 5 No 35 No
David Pryor, AR-D 1 Yes 70 Yes
Donald W. Riegle Jr., MI-D 1 Yes 80 Yes
John D. Rockefeller IV, WV-D 1 Yes 70 Yes
William V. Roth Jr., DE-R 5 No 45 No
Malcolm Wallop, WY-R 5 No 5 No
Sources: NFIB ratings from author's interviews, ADA ratings from Barone and Ujifusa (1993). Cosponsor and vote in
committee come from Legislate on-line service.
Table 5.13. Member Targets in Senate Finance Committee
Senator Number of Groups Mobilizing

John Breaux, LA-D 11
Max Baucus, MI-D 9
Kent Conrad, ND-D 8
John Chafee, RI-R 8
David Boren, OK-D 5
Bill Bradley, NJ-D 5
Daniel Patrick Moynihan, NY-D 3
Total 13
Chafee and Breaux were considered two especially important targets be-
cause of their influence with two important Senate constituencies: moderate
Republicans and conservative Democrats.
Table 5.13 reports targeting choices of interest groups. Of the sixteen
groups who targeted the Finance Committee, targeting information was
available from thirteen.
Once again, the actual targeting decisions were consistent with expecta-
tions. The undecided and influential were more likely to be targeted by
groups attempting to stimulate constituent communications to Congress. In
addition, consistent with the general notion that groups employing grass
roots tactics with a legislative objective should target their efforts where they
have the greatest probability of making a difference, many groups pursued a
small-state strategy. As one lobbyist put it, "fifty small businessmen are
going to influence Max [Baucus] in Montana a lot more than fifty small
businessmen are going to influence Moynihan in Manhattan." In other
words, the lower the denominator - the fewer constituents - the higher the
probability that any one communication would make a difference. Similarly,
advertising time was cheaper in Montana and other small cities.
Also, in the Senate, a group known as the mainstream coalition - or the
rump group - comprised a virtual sixth committee dealing with health care
legislation. Comprised of conservative Democrats and moderate Republi-
cans in the Senate, and under the tutelage of Senator Chafee (also a member
of Senate Finance) the mainstream coalition worked to forge a compromise
plan. Unofficial members included Republicans Durenberger, Danforth,
and Mark Hatfield of Oregon as well as Democrats Joseph Lieberman of
Connecticut, Sam Nunn of Georgia, Byron Dorgan of North Dakota, Frank
Lautenberg of New Jersey, J. Bennett Johnston Jr. of Louisiana, and Bob

Kerrey and James Exon of Nebraska (Clymer et al. 1994; Congressional
Quarterly Almanac 1994, 351; Johnson and Broder 1996).
These senators held the key to breaking a filibuster (Brady and Buckley
1995). Moreover, many of them were considered cue givers. For instance,
Sam Nunn's major area was defense issues, but he was also very influential
with conservative southern Democrats. Laurie Sullivan, an Aetna vice-
president, described her company's rationale for targeting Lieberman: "His
network of contacts gives him greater influence than many other senators"
(quoted in McDonald 1994). Although my data do not permit me to mea-
sure the level of interest group attention to this other virtual committee,
comments and media accounts indicate that there was significant grass roots
attention paid to the rump group.
Chapter Summary
In 1993 and 1994 millions of Americans communicated their feelings about
health care to Congress. This chapter has demonstrated that to understand
why people contacted Congress and what influence these contacts had, we
must understand the elite decisions driving much of this participation. In
short, this chapter has demonstrated that whether we want to understand
mass participation or interest group influence, we must understand how
institutional and political forces structure lobbying choices. In the battle
over health care reform, demographics and attitudes alone did not deter-
mine who participated, and PAC donations and inside lobbying were not the
only ways that organized interests attempted to influence the process.
The predominant strategic objective of interest groups pursuing grass
roots tactics in the battle over health care reform was to influence the cur-
rent legislative battle by demonstrating constituent opinion and by estab-
lishing traceability. Given this objective, the choices and decisions around
health care were consistent with the theoretical expectations outlined in
Chapter 3. Admittedly, my data do not comprise a representative sample of
all choices surrounding health care lobbying. Still, a diverse set of evidence
demonstrated that interest groups carefully targeted their grass roots efforts
at the districts of persuadable legislators on key committees and at influen-
tial members in the Congress as a whole. Within these chosen districts,
employees, customers, suppliers, and stockholders were targeted with mes-
sages that accentuated the short-term costs of health care reform.
To reiterate, I am not suggesting that there were not other objectives -
mobilizing and/or rewarding friends, for example - or that other members of
committees and other members of Congress were not the targets of other
types of lobbying. Consistent with my framework of lobbying, interest
groups possess multiple motives and multiple ways of achieving their goals.
Interest groups may well have been trying to get supporters to the table and/
or encouraging them to work harder once there.
For example, the top five recipients of contributions from PACs with an
interest in health care were John Dingell ($170,460), Marjorie Margolies-
Mezvinsky ($143,860), Lynn Schenk ($111,524), Thomas Bliley
($104,950), and Craig Washington ($98,100). The targeting patterns of
these contributions were not the same as the targeting choices for grass roots
campaigns. Yet, these data are not a competing piece of evidence. Instead,
they demonstrate that scholars striving to understand interest group tactics
must understand how tactics are influenced by strategic objectives and must
be precise about what these strategic objectives are.
Finally, since observations and interviews demonstrated that interest
groups mobilized in expected ways during the battle over health care, the
logical next question is whether these mobilization decisions filtered
through to the mass public. Is there evidence that mobilization mattered and
that it was pursued in the expected places? In the next chapter, in an effort to
answer this question, I combine the information I gathered on targets in this
chapter with two national public opinion surveys conducted during the
summer of 1994.
Patterns of Recruitment and Participation in
the Mass Public
In the previous chapter, I used formal interviews, media accounts, and my

own observations to gather information from the top down on the lobbying
choices and targeting decisions of interest groups in the battle over health
care reform. Consistent with theoretical expectations, the evidence revealed
that the legislative targets in grass roots campaigns on health care were likely
to be undecideds, key committee members, and cue givers. Greater varia-
tion existed among constituency targets - with some groups concentrating
on members, employees, and stockholders, while other groups pursued
more broad-based tactics. Still, the evidence indicated that lobbyists tar-
geted more engaged and influential constituents who could be counted on
to deliver the correct messages.
In this chapter, I investigate from the bottom up the impact of these
decisions at the mass level. I investigate whether there is evidence among the
mass public to confirm the reports of lobbyists and support my model of
mobilization choices. Were citizens residing in states represented by un-
decideds and cue givers on key committees more likely to be contacted by
interest groups? Were more educated and wealthy citizens more likely to be
contacted? Were contacted citizens and those living in targeted districts and
states more likely to get in touch with Congress about health care legislation
or the Clinton plan?
The fundamental approach in this chapter is to add contextual informa-
tion about targeted districts to individual-level survey data containing mea-
sures of participation and mobilization.1 This method allows me to vary the
strategic situation and gauge how the political environment surrounding
health care influenced both elite mobilization and citizen participation. The
1
I had also hoped to merge data on targeted districts with information on patterns of incoming
communications from individual legislator's offices. However, as discussed in more detail in
Appendix B, low response rates to the congressional communications survey I conducted
and concerns about the reliability and validity of the data precluded me from using it for
more in-depth analysis.
106
Recruitment in the Mass Public 107
information on targeted states and districts comes from the research re-
ported in Chapter 5. The survey data come from two national surveys
conducted during the summer of 1994.
Stimulating Participation in Targeted States

First, I use a 1994 Battleground Poll to study the direct effect of recruitment
contacts and to examine recruitment patterns for health care reform lobby-
ing directed toward the Senate (see Appendix A). The Battleground Poll
was conducted during August 1994. It questioned a national sample of one
thousand registered voters and comprised a joint effort between two Wash-
ington, D.C., polling firms: a Democratic firm, Mellman-Lazarus-Lake,
and a Republican firm, The Tarrance Group.
Admittedly, the Battleground Poll is not ideal for scholarly purposes. It
only questioned registered voters who, by definition, have a predisposition
toward participating in politics. It also had a short field-period of only four
days, raising questions of nonresponse bias. Moreover, the "communicating
to Congress on health care" question was asked of only half the sample. And
finally, interest group affiliations as well as certain useful demographic and
attitudinal questions such as church attendance, religion, and political
efficacy were not included.
Nevertheless, callback procedures were followed, professional, well-
trained interviewers were used, and extensive monitoring was employed. It
was also one of the few surveys to include questions about communications
to Congress during the battle over health care reform. Furthermore, it is the
only 1994 survey of which I am aware that asked questions about being
recruited to communicate with Congress and about communications to
Congress on the specific issue of health care. The instrumentation on health
care-related participation allows me to combine these data with information
on the legislative targets I gathered and reported in Chapter 5.
The Model
As discussed in much greater detail in the preceding chapter, the main
targets for players in the health care battle were undecided members of three
key committees - Senate Finance, House Ways and Means, and House
Energy and Commerce. In addition, a group of moderate senators, dubbed
"the rump group," comprised a virtual fourth targeted committee and were
also the focus of grass roots mobilization campaigns. Although health care
legislation was debated in two other committees, neither opponents nor
proponents of the Clinton health care plan paid much attention to the
Senate Labor and Human Resources Committee and the House Education
Table 6.1. Targeted States
States Senators
Connecticut Joseph Lieberman (D)
Georgia Sam Nunn (D)
Louisiana John Breaux (D) *, J. Bennet Johnston Jr. (D)
Montana Max Baucus (D) *
Nebraska Bob Kerrey (D), James Exon (D)
New Jersey Bill Bradley (D) *, Frank Lautenberg(D)
North Dakota Kent Conrad (D) *, Byron Dorgan (D)
Oklahoma David Boren (D) *
Rhode Island John Chafee (R) *
Tennessee Rep. Jim Cooper (D)
Kansas Rep. Jim Slattery (D)
* Member of Senate Finance Committee.

Source: Author's compilation. See text.
and Labor Committee. Both of these committees had strong liberal major-
ities and were unlikely to provide much useful information and political
intelligence to their respective bodies.
There were indications of other grass roots campaigns directed toward
undecided members in the full bodies. Nevertheless, in both houses, the
major focus of mobilization efforts was at the committee level. Unfortu-
nately, the Battleground Poll only recorded the state of each respondent and
not his or her Congressional district. Therefore, I was only able to combine
the Battleground data with information on targeted senators and targeted
states. In the following section of this chapter, I use a Times-Mirror poll to
examine participation patterns in the House (see Appendix A).
The data from my elite interviews revealed that nine states were targeted
for grass roots mobilization campaigns around health care. Extensive state-
wide mobilization campaigns were also conducted in Kansas, where Repre-
sentative Jim Slattery was running for governor, and in Tennessee, where
Representative Jim Cooper was running for the Senate. The full list of
targeted states is contained in Table 6.1.
There was not a sufficient number of respondents in the Battleground
Poll to examine behavior in the individual states. But the responses of resi-
dents of targeted states can be combined to provide a contextual variable
that measures a similar political environment. As described in more detail in
the previous chapter, this atmosphere was characterized by extensive radio
and television media campaigns, mailings, organized town meetings, profes-
sional phone banks, and grass roots campaigns run by full-time state or-
ganizers. Accordingly, the survey question on whether a respondent was
personally urged to participate measures direct mobilization. The targeted
state variable measures other indirect mobilization influences and the effect
of being in a competitive political environment.
In addition to being contacted by a group and living in a targeted state, I
also include other common correlates of participation that should be a part
of any well-specified model of communicating with Congress.2 These
include age, race, education, marital status, employment status, self-
described class, and party identification. (See Appendix A for a description
of question wordings and codings.)
I present these data in both bivariate and multicausal form. For the multi-
causal case I estimate my models with logit regression. Due to the dichoto-
mous nature of the dependent variable, logit regression makes good meth-
odological sense and has often been used when models of participation are
estimated. Logit regression makes good theoretical sense as well. We should
not expect mobilization to mean the same thing to different people at
different times. Mobilization should mean the most to those on the cusp -
and this is precisely what the S-curve of the logit estimator assumes.
To gauge the independent influence of living in a targeted state on being
recruited by a group or lobbying firm, I also model the process by which
citizens are asked to communicate to Congress. Unfortunately, although the
Battleground Poll did ask if respondent were mobilized to contact Congress,
it does not ask what they were mobilized about or who mobilized them.
Consequently, the dependent variable in this model is measuring general
mobilization, while the contextual independent variable measures the politi-
cal environment just about health care. Nevertheless, with the enormous
attention and resources paid to the targeted legislators in the battle over
health care, even with this source of measurement error, we still should
expect to find evidence that the political environment surrounding health
care reform influenced even general patterns of mobilization.
Results
According to the Battleground Poll, one in five registered voters (20 per-
cent) reported that he or she had communicated with a member of Congress
about health care in the past year, and more than four in ten (43 percent)
2
As noted previously, the survey did not contain many of the questions that scholars com-
monly have used in modeling participation - church attendance, efficacy, length of resi-
dence, political knowledge, media habits, or home ownership. In modeling the likelihood of
being mobilized it would also have been useful to have some measure of interest group
membership. Still, even though recruitment of members is a common tactic, organized
interests often cast their net in a wider fashion as well.
Table 6.2. Personal Characteristics and Communicating with Congress on

Health Care
Higher Than Average Levels of Lower Than Average Levels of

Contacting Contacting
White: 21% Black: 15%
Age 65 and Over: 30% Age 18-34: 11%
College Grad: 30% High School Grad: 13%
Republican: 22% Democrat: 18%
Upper Class: 38% Working Class: 13%
Working: 25% Not Working: 17%
Married: 25% Nonmarried: 12%
Mobilized: 36% Not Mobilized: 10%
Targeted State 32% Nontargeted State 18%
Note: N = 500. For codings and question wordings, see appendix A.

Source: Battleground Poll, August 1994.
reported that they had been mobilized. As I noted before, the mobilization
question did not ask on what issue a respondent was mobilized. Also, a little
over one-third (34 percent) of registered voters in the Battleground Poll
reported they had telephoned, faxed, or sent a letter in the past year on any
topic. Table 6.2 provides more information on exactly who was likely to
convey their opinions on health care issues to Congress.
Consistent with previous studies of participation, those more likely to
contact Congress were the more educated, the employed, the married, and
the wealthy. It was also the case that Republicans were more likely to com-
municate their feelings on health care to their representatives in Wash-
ington. (More on this point later in the chapter.) In addition, there was a
slight racial gap, with blacks less likely to voice their feelings about health
care than whites.
Interestingly, members of union households communicated on the health
care issue at exactly the same rate as the population as a whole. This was not
the case with communications to Congress in general, where members of
union households out-participated nonunion households by 16 percentage
points (41 percent to 25 percent). This finding is consistent with what I was
told in my interviews - that unions had exhausted most of their time and
resources on NAFTA and never became a player in the battle over health
care.3
3
I should note that one union official whom I interviewed angrily denied the claim that unions
never showed up for the battle over health care.
Mobilized in Not Mobilized in Not

Targeted Mobilized in Non- Mobilized in
State Targeted targeted Non-
State State targeted
State
Figure 6.1. Mobilization and participation in target and nontarget states (n =
1,000). For coding and question wordings, see Appendix A. Source: Battleground
Poll, August 1994.
Moving to the main topic of interest, respondents who resided in any of

the eleven states targeted for statewide grass roots campaigns in the battle
over health care had a participation rate 14 percentage points higher than
those in other states (32 percent to 18 percent). Looking at the effect of
direct mobilization on communications to Congress about health care, more
than one-third (36 percent) of those who claimed a mobilization contact
reported communicating to one of their senators or representatives in Wash-
ington about health care. In sharp contrast, only one in ten (10 percent) of
noncontacted respondents reported such communication.
Figure 6.1 combines information about direct recruitment and the politi-
cal environment. Specifically, it illustrates the participation rates for mobi-
lized and nonmobilized respondents in targeted and nontargeted states.
As the bar chart in Figure 6.1 illustrates, citizens receiving a recruitment
contact in targeted states were more likely to participate than citizens receiv-
ing a recruitment contact in nontargeted states. Similarly, albeit at a much
lower level, citizens receiving a recruitment contact in nontargeted states
were more likely to contact Congress than noncontacted citizens in non-
targeted states. Not surprisingly, citizens receiving a recruitment contact in
targeted states had the highest level of contacting Congress (50 percent).
Looking at the effect of recruitment activities in yet a different way, almost
three in four persons who contacted Congress about health care legislation
(72 percent) reported that they had been asked to do so. Table 6.3 illustrates
exactly which citizens were likely to receive a mobilization contact. Even
Table 6.3. Personal Characteristics and Being Mobilized to Contact Congress

Mobilization Mobilization
Male: 46% Female: 39%
Age 65 and over: 46% Age 18-34: 29%
Married: 46% Nonmarried: 38%
Working: 25% Not Working: 17%
Union Household: 51% Nonunion Household: 41%
Upper Class: 50% Working Class: 22%
Republican: 44% Democrat: 40%
Target State: 50% Nontargeted State: 40%
Notes: N = 1,000. For codings and question wordings, see Appendix A.

using this broad measure of recruitment, respondents in the eleven states

targeted for health care lobbying were more likely to receive a request to
participate. Also, as expected, the educated and the self-described upper
class were more likely to have received a recruitment contact. Furthermore,
there was a slight gender gap, with men more likely to receive a request than
women. Republicans were also slightly more likely to be recruited than
Democrats.
So far, at least in the bivariate case, there is evidence that elite efforts
mattered and that lobbyists targeted influential constituents to communi-
cate with influential members of Congress.
Before presenting the results of the logit regression, I must address a
tricky theoretical and statistical problem extant in studying the effect of
recruitment contacts on participation. Put simply, do recruitment efforts
cause citizens to participate, or are citizens targeted because they are likely
to participate in the first place? Statistically speaking, in determining the
effect of a mobilization contact, there is a selection bias due to nonrandom
assignment. Such a problem arises when methods assume a random assign-
ment to different treatment groups when, in reality, this is not the case. A
model that wrongly assumes a random assignment will yield a stochastic
component with an expected value not equal to zero.
Since the point of this project is to argue that elite lobbying decisions are
far from random and have a unique political logic, this is clearly a potential
Table 6.4. Causes of Phoning, Faxing, or Writing Congress on Health Care
Standard Partial
Variable Coefficient Error Effect
White -.21 .37
Male -.01 .25
Under 35 -.78 .49
Over 65 .73 .30 8
Education 1.14 .43 13
Strong Partisan .05 .29
Union Household .15 .32
Employed .70 .22 8
Upper Class .98 .38 11
Republican .33 .22
Married .80 .29 10
Targeted State and Not Mobilized .77 .33 9
Mobilized 1.40 .25 20
Notes: N = 500. Percentage of cases correctly predicted = 82% (null mode =

80%). For codings and question wordings, see Appendix A.
problem. In fact, part of the logic of my theoretical model is that citizens are
being contacted for a reason that is tied to their probability of participating.
In practice, however, I think the consequences for my estimates are quite
small. My model already includes many of the factors that lobbyists might
employ in making their recruitment choices. Also, the sample comprised
registered voters who were reachable in a four-day period by phone. These
are precisely the types of citizens that lobbyists are likely to target - and are
likely to be able to target. These conditions may bias in an upward fashion
the logit coefficient measuring mobilization contacts (the effect of mobiliza-
tion might not be so high for the general public). But it also lessens the
chance that a coefficient might be biased due to nonrandom selection.
Moreover, it is unlikely that interest groups targeted whole states with a
predisposition to participate and that those states precisely matched my set
of lobbying targets.
The findings from the bivariate analysis are confirmed when a logit model
is used to sort out individual effects. The results of the model are reported in
Table 6.4. In addition to the coefficients and standard errors, for those
coefficients that were statistically significant at the commonly used 95 per-
Table 6.5. Causes of Being Mobilized
Standard Partial
Variable Coefficient Error Effect
White .27 .19
Male .28 .14 7
Under 35 -.76 .22 -15
Over 65 .32 .17
Strong Partisan -.06 .18
Employed .28 .13 7
Upper Class 1.35 .26 27
Republican .34 .29
Union Household .40 .17 9
Married .16 .15
Targeted State .48 .20 11
Notes: N = 1,000. Percentage of cases correctly predicted = 64% (null

model = 57%). For codings and question wordings, see Appendix A.
cent confidence level, I also report the partial effects of independent vari-
ables (in percentage points) on the probability of communicating with Con-
gress. For each of the variables, I calculated the probability that each respon-
dent would communicate to Congress under two scenarios, assuming that
the variable takes its lowest value and then assuming that it takes its highest
value, allowing all other variables to take their observed values. The re-
ported effect is the difference between the two probabilities averaged across
the entire sample.
With other factors held constant, recruitment still had a large effect on
participation rates. Directly recruited individuals were 20 percentage points
more likely to communicate to Congress than those who were not contacted
by an interest group.
Looking at the individual attributes that influenced who was contacted,
Table 6.5 reports the results of a model in which being asked to communi-
cate with Congress was the dependent variable. With other influences held
constant, we see that whites, males, and the better educated were signifi-
cantly more likely to receive a request. Younger voters were less likely to be
targeted for recruitment.
This analysis provides more evidence that lobbyists targeted individuals
for whom mobilization would make a difference in their decision to partici-
pate as well as individuals whose communications would be influential.
Most importantly, we also have further confirmation that citizens who re-
sided in states represented by influential senators were more likely to be
targeted for a contact. With all else held constant, a resident of a targeted
state was 11 percentage points more likely to be targeted.
Because of their organizational affiliation, members of union households
were more likely to be contacted. As discussed previously, however, unions
never really came to the fight in the battle over the Clinton health care plan.
Although members of union households were more likely to receive a gen-
eral recruitment contact, they were no more likely to contact their legislators
on health care legislation. But, on the general communicating to Congress
question, by a margin of 41 percent to 31 percent, members of union
households were more likely to communicate with Congress. This finding
holds up in a multicausal model as well (Goldstein 1995).
To examine further the characteristics of constituents targeted for mobili-
zation, I calculated a set of representation ratios. A representation ratio
measures how underrepresented or overrepresented a particular group is
relative to its share of the total population. For example, if college-educated
citizens comprise 20 percent of the potential electorate and 30 percent of the
actual electorate their representation ratio would be 1.50. Likewise, if lower-
income individuals comprise 30 percent of the potential electorate but only
15 percent of actual voters, their representation ratio would be .50 (see
Rosenstone and Hansen 1993).
As Table 6.6 illustrates, the focus of lobbyists on higher-end constituents
is also demonstrated by the preceding representation ratios. They show that
college graduates and the self-described upper class were contacted by
groups at a rate greater than the proportion they composed in the popula-
tion as a whole. In general, these findings about recruitment patterns and
citizen communications to Congress underscore the fact that grass roots
lobbying tactics create multiple biases in favor of those with greater re-
sources. Those with greater resources and skills are not only more likely to
have the ability to communicate with Congress in the first place, but are also
more likely to be contacted by elites. Further, their participation is likely to
count more with members of Congress.
To sum up, according to the Battleground Poll, recruitment had a statis-
tically and substantively large influence on individuals' likelihoods of con-
tacting Congress on health care. Even with all the national attention paid to
the issue and the personal stake that every American had in the legislation,
citizens still needed a push to participate. Moreover, there was evidence at
the individual level that lobbyists systematically targeted their efforts where
they were most likely to influence the ultimate outcome. Residents of states
with undecided or influential senators on key committees were more likely
Table 6.6. Representation Ratios for Being Mobilized
Subgroup Representation Ratio

Education
Less than High School .7
High School .75
Some College 1.03
College Grad 1.28
Class
Working Class .72
Lower Class .78
Middle Class 1.10
Upper Class 1.65
Note: N = 1,000.
to be targeted and were more likely to contact Congress. Also, there was
evidence that organized interests targeted their recruitment efforts toward
those who were likely to respond and whose contact would have an influ-
ence. Respondents with higher resources and more influence were more
likely to be contacted and were more likely to communicate with Congress.
In short, the behavior reported in the Battleground Poll is consistent with
the theoretical expectations outlined in Chapter 3 and the empirical findings
regarding interest group recruitment choices reported in Chapter 5.
Mobilization and Targeted House Districts

Turning to another data set, I use a July 1994 Times-Mirror poll to examine
participation focused on the House (see Appendix A). The study does not
have self-representing samples of congressional districts. Still, the large
number of respondents (3,800 voting-age population) ensures an adequate
representation when House districts with similar political environments are
aggregated.4
4
The Times-Mirror Poll does not ascertain the congressional district of respondents. How-
ever, the staff at the Times-Mirror Center for the People and the Press (now the Pew Center)
provided me with the telephone numbers of respondents. With the phone numbers, I worked
with Survey Sampling, a Connecticut-based sampling company, to match the area codes and
exchanges to particular congressional districts. Because of some overlap with exchanges and
congressional districts, this process was not perfect. In the 11 percent of the cases where
exchanges overlapped districts, the district where the majority of the exchange was located
was selected. There is no reason to believe that this introduced any sort of systematic bias.
Although the Times-Mirror questionnaire contains more of the variables
that should be correlated with participation (political information, church
attendance, efficacy, and media usage) than does the contemporaneous
Battleground Poll, it does not ask whether respondents were recruited and
does not ask respondents why they contacted Congress. These omissions
pose serious problems. For instance, the lack of a question on recruitment
makes it impossible to examine the direct effect of a recruitment contact.
Furthermore, since different legislators and Congressional districts were in
play with different issues, not knowing which issue spurred a respondent to
contact his or her legislator makes it difficult to add contextual data on the
political environment in the district.
Nonetheless, enormous attention was paid to lobbying over health care.
Even using a general measure of participation, we should still expect to find
higher rates of participation in targeted districts. Furthermore, since there
was a major push to spur constituent communications on health care in the
spring of 1994, a general question asked immediately after a flurry of mobili-
zation activity on health care in July 1994 should pick up variations in areas
targeted. Therefore, even though the Times-Mirror poll did not measure all
the different types of grass roots campaigns that were conducted, the district
targeting variable I constructed used the information on House targets in
the battle over health care reform that I gathered from my interviews dealing
with health care reform.
The roster of targeted members and districts is presented in Table 6.7.
The targeted districts are the ones that were represented by swing members
of the House Energy and Commerce Committee as well as the House Ways
and Means Committee. The information comes from the empirical work
reported in Chapter 5.
In modeling contacting Congress, I included the following additional
variables: age, race, sex, education level, employment status, member of a
union household, marital status, home ownership, church attendance,
strength of partisanship, party identification, political knowledge, political
efficacy, and media habits. Again, I present the data in both bivariate and
multicausal form and estimate my multicausal model with logit.
Results
A little over one-quarter (28 percent) of respondents to the July 1994
Times-Mirror Poll reported that they had contacted their senator or repre-
sentative in Washington during the past year. Table 6.8 provides the bivari-
ate breakdown of those who reported communicating to Congress in the
past year.
Again, consistent with the Battleground Poll's findings on health care
participation and scores of previous studies on the correlates of participa-
Table 6.7. Targeted House Districts
Member Party State and District

Slattery Democrat KS-2
Cooper Democrat TN-4
Boucher Democrat VA-9
Lehman Democrat CA-19
Schenk Democrat CA-49
Margolies-Mezvinsky Democrat PA-13
Lambert Democrat AR-1
Tauzin Democrat LA-3
Klug Republican WI-2
Upton Republican MI-6
Hall Democrat TX-4
Hoagland Democrat NE-2
Brewster Democrat OK-3
Payne Democrat VA-5
Jefferson Democrat LA-2
Neal Democrat MA-2
Kennelly Democrat CT-1
Johnson Republican CT-6
Levin Democrat MI-12
Source: Author's compilation. See Chapter 5.
tion, those most likely to contact members of Congress were whites, married
couples, homeowners, those with higher levels of political information, sen-
ior citizens, and the more educated. Conversely, blacks, younger citizens,
and the poorly educated were less likely to participate. Union households as
well as Republicans were also more likely to have communicated with their
members of Congress.
Notably, the findings on the behavior of talk radio listeners are dramatic.
Talk radio listeners contacted Congress at a rate twice that of the population
as a whole (39 to 20 percent). "Dittoheads," as Rush Limbaugh's listeners
call themselves, participated at a rate 16 percentage (42 percent to 26 per-
cent) points higher than "non-Dittoheads."
Finally, there is also evidence for recruitment contacts and the effect of a
competitive political atmosphere. As Table 6.8 illustrates, respondents re-
siding in districts of targeted members of the House Energy and Commerce
Committee and the Ways and Means Committee contacted Congress at a
Table 6.8. Individual Characteristics and Communicating to Congress

Contacting Contacting
Over 35: 32% Age 18-34: 18%
Union Household: 32% Nonunion Household: 27%
Male: 29% Female: 27%
Married: 32% Nonmarried: 23 %
Homeowner: 32% Nonhomeowner: 20%
Frequent Church Attendee: 35% Non-Church Attendee: 25%
High Political Info: 36% Low Political Info: 17%
Strong Partisan: 32% Independent: 25%
Talk Radio Listener: 39% Nonlistener: 20%
Limbaugh Listener: 42% Nonlistener: 26%
Targeted Congressional District: 35% Nontargeted Congressional
District: 28%
Note: N = 3,800. For codings and question wordings, see Appendix A.

Source: Times-Mirror survey, July 1994.
rate 7 percentage points higher than in nontargeted districts. Again, due to

limitations in the survey instrument, I am unable to include direct recruit-
ment in this model.
Examining the effect of these variables with other factors held constant,
Table 6.9 presents a summary of the results from a logit model. Once again,
I report the coefficients, standard errors, and the statistically significant
coefficients transformed into probabilities.
We see that the more educated, whites, those who were married, Republi-
cans, and members of union households were the most likely to contact
Congress. Younger voters were less likely to participate. Most notably, with
other important correlates of participation held constant in the model, con-
stituents of undecided members of the Ways and Means Committee and the
Energy and Commerce Committee were 6 percentage points more likely to
contact Congress.
The mobilization power of talk radio and Limbaugh listeners also shines
through in the multicausal analysis. Frequent talk radio listeners were 12
percentage points more likely to communicate with Congress. This finding
should be treated with caution since listening to talk radio is clearly endoge-
Table 6.9. Causes of Contacting Congress
Standard Partial
Variable Coefficient Error Effects
White .48 .12 9
Under 35 -.46 .11 -8
Union Household .26 .10 4
Male .09 .08
Married .16 .08 3
Homeowner .09 .10
High Political Info .80 .15 14
Strong Partisan .05 .13
Republican .34 .08 6
Efficacy .11 .08
Church .58 .22 8
Talk Radio Listener .75 .10 12
Limbaugh Listener .31 .14 5
Targeted Congressional District .36 .16 6
Notes: N = 3,800. Percentage of cases correctly predicted = 75% (null

model = 72%). For codings and question wordings, see Appendix A.
nous to participating in politics. Nevertheless, this finding is consistent with

other information I happened across on the influence and use of talk radio
by political parties and interest groups.
Newt Gingrich's office and the Republican National Committee were
especially aggressive about faxing talking points and legislative alerts to
various high-profile talk radio hosts during the 103rd Congress. Speaking
about a local talk show host, a Congressional staffer told me, "You knew it
when he mentioned our number because suddenly the phones would light
up right after he went on the air." A Limbaugh producer was even present at
the daily meetings of an ad hoc coalition found to oppose the Clinton health
care plan. The National Rifle Association (NRA) was especially aggressive
about getting its message out over talk radio. Furthermore, although the
Democratic Party was a little behind the GOP, both parties now have offices
and full-time staff members responsible for tracking talk radio and acting as
a liaison to the various talk radio hosts. The Democratic House Caucus even
brought in an expert on talk radio to its annual retreat. Although I cannot
tackle the topic of talk radio and mass participation in this project, it surely
deserves further attention. At minimum, it is more evidence that politics
matters and that demographic characteristics alone do not explain par-
ticipation.
One last way that I test the effect of politics on participation is to take
advantage of the fact that the Times-Mirror Poll asked a two-part question
on contacting Congress. The first part asked whether a respondent had
"ever" contacted Congress and the second asked whether a respondent had
contacted Congress in the "last year." Since the information I had on tar-
geted districts was from a recent lobbying campaign (health care), we should
not expect living in a targeted state on health care to influence a respon-
dent's previous history of participation. In other words, although the effect
of resources and attitudes should stay the same, living in a state targeted for
health care lobbying should have no effect on previous patterns of participa-
tion. To test this hypothesis, I used the same model as used previously,
substituting "contacting Congress ever," for "contacting Congress in the
past year" as the dependent variable. As expected, and providing strong
confirmatory evidence for my model, living in a targeted state for health care
has no effect on having ever contacted Congress. The effect for living in a
targeted state was .12 with an accompanying standard error of .16.
A Republican Tinge
Finally, in addition to the final legislative outcome - or lack of outcome -
evidence from public opinion surveys suggests that critics of the Clinton
plan were more successful in their efforts to stimulate opposition to the
Clinton plan. As Table 6.10 demonstrates, Republican partisans were more
likely than Democrats to contact Congress in 1994.5 This Republican tinge
to the electorate represented a new phenomenon. Comparing those in the
Times-Mirror Poll who contacted Congress in the past year with those who
contacted Congress in previous years illustrates the Republican bent in
1994 (see Table 6.11).
The results in Table 6.11 are striking. Not only were Republicans and
those inclined to vote for Republicans in congressional races more likely
than Democrats to contact Congress in 1994, but, as shown by the differ-
ences in the past-year versus previous-year measures, the magnitude of the
differential was significantly higher than in previous years. In addition, and
consistent with this finding, Republicans made up a larger proportion of
those who contacted Congress: over half (52 percent) were Republican,
5
The Battleground Poll, taken during the same time period, shows similar results on the
partisan composition of those contacting Congress.
Table 6.10. Contacting Congress and Partisanship
High Levels of Communicating Low Levels of Communicating

Republican: 34% Democrat 23%
Republican Congressional Voter: Democratic Congressional Voter:
33% 18%
Disapprove of Clinton: 34% Approve of Clinton: 23%
Notes: N = 3,697. Cells = proportion of groups contacting a member of

Congress. For coding, see appendix A.
Table 6.11. Proportion of Party Identifiers and Party Voters by Time of

Contact
Party Identification Congressional Vote

Democrat Republican Democrat Republican
Contacted MOC in 23% 34% 24% 33%
past year
Contacted MOC 29% 34% 30% 33%
but not in past
year
Notes: N = 3,697. Cells = proportion of groups contacting a member of

Congress (MOC). For coding, see Appendix A.
while fewer than four in ten (39 percent ) were Democrats. In previous
years, the same number of Democrats and Republicans contacted Congress
(Democrats composed 47 percent of communicators and Republicans com-
posed 46 percent of communicators).
Using National Election Study (NES) data, Table 6.12 makes this identi-
cal point. Although Republicans out-participated Democrats in 1980 (the
year of Ronald Reagan's landslide victory over Jimmy Carter and a Republi-
can takeover of the Senate) rates of participation mirrored each other in
1986 and 1990. However, similar to the findings in the Times-Mirror Poll,
Republicans in 1994 were more likely to communicate their feelings to their
representatives in Congress.
Table 6.12. Communicating with Congress to Express Opinion on Issues,

1980-1994
Republican Democrat
1980 6.5% 3.8%
1986 7.1% 6.6%
1990 7.1% 6.3%
1994 12% 6.0%
Note: For coding, see Appendix A.

Source: American National Election Studies.
Evidence from two separate surveys not only demonstrates that Republi-
cans were more likely than Democrats to communicate to Congress before
the 1994 election, but that the GOP advantage was more pronounced than
it had been in past years. While Democratic rates of participation essentially
remained steady, GOP rates enjoyed a huge jump.
Chapter Summary
Politics matters: variations in participation and recruitment patterns among
the citizenry are the result of systematic political choices by elites. Although
more precise information on individual patterns of recruitment and par-
ticipation is clearly needed, the data reported in this chapter are consistent
with the theory of mobilization choices developed in this project and con-
firm the self-reports of the grass roots lobbyists I interviewed.
As demonstrated in Chapter 5, the legislative targets of grass roots lobby-
ists were likely to be undecideds, key committee members, and cue givers.
Within these targeted states and districts, the survey data indicated more
engaged and influential constituents were more likely to be recruited. Those
citizens who lived in the districts of legislators who had a high probability of
influencing the process - and who themselves had a high probability of
influencing that legislator - were more likely to be recruited and to contact
Congress.
In addition, the data presented in this chapter provide evidence that
recruitment matters and that recruitment choices need to be part of any
correctly specified model of mass participation. Any analysis of participa-
tion patterns during the fight over health care reform that concentrated only
on resources and social-psychological attitudes would have painted an in-
complete picture of the causes of participation. The results presented in this
chapter demonstrate that the logic of mobilization at the individual level

reinforces the biases that exist in who participates.
Finally, this chapter demonstrates that studies that have attempted to look
for evidence of interest group behavior at the national level in the battle over
health care reform missed an important point. The goal of interest groups
was to influence particular segments of the public in particular constituen-
cies. It was to influence particular members of Congress, and in no small
way by activating and mobilizing particular segments of their constituencies.
The goal was not to move national survey numbers or to convince every
member of Congress.
7
Conclusion
Grass roots mobilization is used for one purpose,

period - to influence legislative policy. It's not
about getting more Americans involved. It's not
about educating people on the issues. It's not about
making Americans feel good about their political
system.
Trade association executive speaking at
a workshop on grass roots mobilization
Proportionately, more Americans than ever are exercising their right to
lobby and communicate their feelings to their representatives in Congress.
Such high levels of citizen participation may conjure up images of ordinary
Americans spontaneously organizing to voice their concerns to policy
makers. Furthermore, the fact that Americans are writing, calling, tele-
gramming, faxing, and even e-mailing their representatives more than ever
might seem like grounds for celebration to those concerned with low rates of
electoral participation.
This book, however, has demonstrated that the ringing of phones, the
flurry of letters, and the whirring of faxes is not necessarily evidence that
Main Street is talking to Washington and that all is well with participatory
democracy in the United States. Recent events, such as the National Feder-
ation of Independent Business's activity against the employer mandate in
the Clinton health care plan, the American Energy Alliance's struggle
against the BTU tax, the American Association of Retired Persons' fight
against Medicare cuts, and the National Rifle Association's tactics in op-
position to the crime bill demonstrate that little about mass participation in
the United States is spontaneous. Rather, interest groups and lobbying firms
inside the beltway are increasingly utilizing new and sophisticated tech-
nologies to water the grass roots outside the beltway.
This apparent growth in both grass roots lobbying and communications
to Congress was the inspiration for this project. Working from the premise
125
that a better understanding of lobbying tactics was required in order to

gauge the true nature of interest group influence and the quality of represen-
tation in the United States, my goal was to understand better the use of grass
roots campaigns as a lobbying tactic. In this concluding chapter, I briefly
summarize the main findings of the project, discuss their normative implica-
tions, and propose some modest reforms.
Results
The logic of grass roots lobbying decisions is rooted in the logic of congres-
sional elections and congressional decision making. More precisely, the
informational needs of voters and legislators provide strategic opportunities
for lobbyists to influence both electoral and policy battles. Grass roots lob-
bying is a way for interests groups to provide information to voters and
legislators that can influence election outcomes or the fate of a particular
piece of legislation.
Given a particular type of information that needs to be communicated,
grass roots lobbying decisions are driven by the probability that an individ-
ual's communication will influence a given strategic objective. Whether it
was an intensive look at one case or a more general examination of lobbying
choices across a wide range of issues, interviews with interest group repre-
sentatives, surveys of the mass public, and my own observations provided
evidence consistent with this logic.
As expected, when the objective was a short-term legislative one and the
goal was to provide information to legislators about the consequences of
their actions, lobbyists were likely to target districts represented by un-
decided and influential legislators. For example, in the battle over health
care reform, the targets were swing members on the House Energy and
Commerce Committee, the House Ways and Means Committee, and the
Senate Finance Committee. In the fight over NAFTA, attention was
focused on the districts of undecided representatives in the House as a
whole. In the battle over the BTU tax, attention was focused on members of
the Oklahoma congressional delegation who were important and influential
members of important and influential committees. Within targeted districts
on these as well as other grass roots lobbying campaigns with a legislative
objective, lobbyists made great efforts to target those citizens who had the
best chance of influencing their members of Congress. Although these con-
stituents were often wealthy or well connected - executives, longtime sup-
porters, and neighbors - credibility occasionally demanded more nontradi-
tional messengers, such as waitresses, Boy Scouts, and senior citizens.
When the goal was a more long-term electoral one and the objective was to
provide information to voters and potential challengers, targets were likely to
Conclusion 127
be legislative opponents who were judged to be vulnerable. The Democratic
targets for Medicare lobbying in 1995 and 1996 were Republican freshmen
who had won by narrow margins in 1994 and who represented a large
proportion of elderly constituents. Republican targets in 1994 during debate
over the crime bill were members who represented districts that had tradi-
tionally gone Republican in presidential races. Furthermore, with an elec-
toral objective, I found that constituent targets were likely to be more broad-
based and included those citizens who could most credibly frame the issue.
Whichever the goal, messages were usually crafted to highlight early-order
direct costs and mobilization efforts were generally targeted at those citizens
most likely to bear early-order costs. When groups had the resources, mes-
sages were crafted and tested with focus groups and surveys. Again, with
both goals, citizens over whom groups had leverage, citizens who were likely
to respond to a request, and citizens who were likely to respond in a way
consistent with an interest group's goals were the most likely targets.
Implications
The great majority of the work on mass participation has concentrated on
voting. Much of this previous work on mass participation has attempted to
explain political activity by personal resources (Verba et al. 1996), partisan
attachments (Abramson and Aldrich 1982; Campbell et al. 1960), political
interest (Neumann 1986; Verba and Nie 1972), political efficacy (Abram-
son 1983; Bennet 1986), and political information (Zaller 1992).
Much of the work on lobbying in the past twenty years has concentrated
on the relatively narrow area of PAC donations. Also, the debate in the
lobbying literature has long revolved around the question of whether groups
are more likely to target their friends or their enemies. Different sets of
scholars at different times have found evidence and devised rationalizations
for why interest groups might want to target supporters (Bauer et al. 1964;
Hall and Wayman 1990; Hayes 1981; Mathews 1960; Milbrath 1963;
Ziegler 1964), opponents (Austen-Smith and Wright 1994), or undecideds
(Ainsworth and Sened 1993; Rothenberg 1992; Smith 1984).
My work suggests that these explanations of mass participation and inter-
est group lobbying are theoretically and descriptively incomplete. What's
more, my work suggests that scholars must tackle the two subjects in tan-
dem. Traditional explanations of mass participation find it difficult to ex-
plain variations and fluctuations in patterns of participation. This is because
they have ignored how the nature of our political system creates incentives
for groups to bring into politics particular individuals in particular places at
particular times. The ways that politics brings people to participate are far
from transitory or unsystematic. They proceed from the very nature of our
political system. As my work confirms, resources, attachments, efficacy, and

information are certainly a big part of the picture and certainly influence
who is recruited to participate in politics. Although findings on the power of
mobilization are not novel, this project explains the logic of mobilization and
demonstrates the importance of understanding the logic of mobilization if
we are to understand the nature of mass participation.
Likewise, the mobilization of constituent communications to Congress
must be a part of any model of interest group influence or lobbying. Even
more importantly, future work on interest group strategies and tactics must
proceed from a proper theoretical understanding of exactly what lobbyists
are trying to accomplish. Although previous scholarship has taught us much
about how interest groups operate, the lack of a theoretical picture of lobby-
ing and its strategic objectives has limited much of this previous work's
explanatory power.
For example, why have some scholars found evidence that groups target
friends, while others have found evidence that they target enemies, and yet
others have found that undecided legislators are likely to be the targets? The
answer is simple. Interest groups do, in fact, sometimes target friends, some-
times target enemies, and sometimes target undecideds. Tactical lobbying
choices depend on the lobbying strategy being used and a group's ultimate
strategic objective. Lobbyists have multiple motives and have multiple lob-
bying tools to achieve these motives. Therefore, not only in order to under-
stand tactical mobilization decisions, but in order to study lobbying and
interest group influence in general, we need to be precise about what interest
groups are trying to achieve.
Lastly, understanding lobbying tactics is also crucial for understanding
how issues and incumbent behavior influence congressional elections. Stra-
tegic and tactical mobilization decisions and mass participation outside the
formal congressional campaign period can tell us much about the nature of
congressional campaigns and congressional elections. Scholars of lobbying
have paid extensive attention to the congressional electoral connection. It is
time for scholars of congressional elections to pay attention to the grass
roots lobbying connection.
In sum, looking from the bottom up with survey data and from the top
down with data from interest group interviews, the empirical findings in this
project demonstrate that requests for political activity matter and that they
deserve a place in any correctly specified model of political participation or
interest group lobbying. Understanding why individuals participate in poli-
tics requires more than an understanding of individual attributes and at-
titudes. Similarly, understanding how and why interest groups lobby re-
quires that one look beyond financial donations and the direct activities of
Washington-based lobbyists.
Conclusion 129
Pathologies and Reforms

Concern about biases introduced by low rates of mass participation and the
supposed power of special interests has created a growth industry for would-
be reformers. Although there is no shortage of proposed solutions that claim
to have found the silver bullet, my findings demonstrate that because the
strategies and tactics of interest groups are imbedded in our politics, reform
will not be easy. I have concentrated on only the tactical aspects of one
lobbying tool - mobilizing constituent communications to Congress. But
my work makes the more general point that reforms need to proceed from a
better theoretical understanding of how lobbying works and how participa-
tion works.
Interest groups stimulate mass participation because it is an effective way
for them to provide information to voters and legislators. The lobbying
strategy of mobilizing constituent communications to Congress did not
create the need for this information. It did not create the fact that legislators
will often base decisions on small, intense minorities, or that legislation with
short-term direct costs to an attentive public and long-term general benefits
to an inattentive public would be difficult to pass. It did not create the reality
that opponents of a policy have the advantage over proponents, or that
people react more intensely in opposition to the reduction of an old benefit
than they do in favor of a new one. The lobbying strategy of stimulating
constituent communications has created none of these realities. It, however,
has exacerbated their effects. It has exacerbated already existing forces to-
ward short-term goals and single interests. It has contributed to a society
that increasingly finds it difficult to formulate solutions to complex policy
questions.
Some might take these findings as evidence that scholars such as Be-
relson, who were concerned about the dangers of too much mass participa-
tion, were correct. To the contrary, the reason why stimulated participation
is so effective in conveying information is because participation rates are low.
In particular, low rates of participation in congressional elections (only
about one-third of the eligible voting-age population typically votes in mid-
term elections) make legislators particularly sensitive to the information
they receive from constituent communications and make them less likely to
pay attention to potential constituencies. Similarly, these same low rates of
participation make grass roots campaigns with an electoral objective more
effective. Narrow constituencies targeted for persuasion can have a large
influence in a low-turnout contest.
The usual solution proposed by political scientists for such problems is
stronger parties. As a card-carrying political scientist, it is easy enough for
me to join the call for more powerful parties that would supposedly mobilize
voters and focus legislators on collective results and collective accountabil-

ity. Nevertheless, despite the discipline's pleas, it is hard to see how parties
will become stronger. Short of the creation of responsible party government,
however, there are steps that can be taken.
First, interest groups and lobbyists should have to report the money they
spend to stimulate constituent communications to Congress. Although a
strange alliance ranging from the National Rifle Association to the American
Civil Liberties Union rose up in opposition to stricter reporting require-
ments for those engaging in mobilization activities, the goal is not to restrict
a group's right to engage in politics. It is to increase the rights of the rest of us
to know who is lobbying. Will such reporting requirements alleviate bias in
our system and stop the need to transmit information? No. Given the fact
that legislators and voters need information, however, the public should
know who is transmitting what information to whom - and how they are
doing it.
Second, since low rates of electoral participation exacerbate the effect of
information-sending strategies like stimulated communications to Con-
gress, efforts should continue to increase rates of voting. The way to limit
the effect of strategies that attempt to convey biased information is to
decrease the importance of that information. The empirical jury is still out
on reforms like motor-voter and vote-by-mail elections. Scholars need to
study these reforms and actively voice their findings to policy makers. Tech-
nological advances have spurred the growth in communications to Con-
gress. Careful attention should be paid to new technologies that could in-
crease rates of voting.
Final Words
The central premise of this project has been that elite recruitment of mass
participation ties together important questions of mass participation and
interest group influence. A theoretical and empirical understanding of the
effect and causes of mobilization decisions must be a part of any complete
picture of interest group influence or mass participation. The central argu-
ment has been that the factors that shape mobilization decisions - and thus
the nature of political participation and interest group influence - are rooted
in the basic logic of our political system and the strategic needs of interest
groups to provide information to citizens and legislators. The theory and
empirical analysis presented in this book should hopefully move us toward a
better understanding of how groups influence congressional decision mak-
ing and why individual citizens participate in politics.
Finally, the basic theory of mobilization and participation outlined and
the initial empirical findings presented have serious consequences for Amer-
Conclusion 131
ican democracy. Democratic theories of representation teach us that the
legitimacy of representative government relies on the mass public to com-
municate its beliefs to elite policy makers. Those individuals and groups
who do not speak, are not heard - they do not affect the selection of our
leaders or the public policy choices they make. Paraphrasing Schatt-
schneider (I960), the logic of mobilization and participation presented in
this project demonstrates that the choir leader has an extraordinary amount
of power and the accent of the heavenly choir depends heavily on the strate-
gic choices that the choir leader is making.
Appendix A
Sources and Coding for Survey Data
The Battleground Poll

Sample: National sample of 1,000 registered voters
Field Date: August 18-21, 1994
Communicated with Congress: "Have you called, or sent or faxed a letter to
your congressional representative or senator to express your opinion on an
issue in the past year?" Coded 1 if yes; 0 if no.
Communicated with Congress about Health Care: "Have you called, or sent or
faxed a letter to your Congressional representative or senator to express your
opinion specifically on health care legislation in the past year?" Coded 1 if
yes; 0 if no.
Mobilization Contact: "From time to time parties and other groups contact
people and urge them to phone or fax or call their senator or representative
to convey their feelings on an issue. Has anyone contacted you and encour-
aged you to get in touch with your senator or representative in Washington?"
Coded 1 if yes; 0 otherwise.
Sex: By observation: Coded 1 if male; 0 if female.
Race: "What is your race? Are you African-American, white, Asian, or some
other?"
White: Coded 1 if white; 0 if anything else.
Marital Status: "What is your current marital status - are you single and
never married, married, separated, divorced, or widowed?" Coded 1 if mar-
ried; 0 if anything else.
Party Identification: "Generally speaking, do you think of yourself as a Re-
publican, a Democrat, an independent or what? (If Republican or Demo-
crat) "Would you call yourself a strong (Republican/Democrat) or not very
strong (Republican/Democrat)?" (If independent, other, or no preference)
"Do you think of yourself as closer to the Republican or Democratic party?"
133
134 Appendix A
Republican: Coded 1 if strong, not strong, or lean Republican; 0 if any-
thing else.
Democrat: Coded 1 if strong, not strong, or lean Democrat; 0 if anything
else.
Intensity: Coded 0 if independent or apolitical; .33 if independent leaning
toward a party; .67 if a weak partisan; 1 if a strong partisan.
Union Household: "Are you or is any member of your household a member of
a labor union or teachers' union?" Coded 1 if yes; 0 if no.
Age: "What is your age, please?"
Under 35: Coded 1 if under 35; 0 if anything else.
Over 65: Coded 1 if over 65; 0 if anything else.
Education: "What is the highest level of education you have completed?"
Coded 0 if 8 grades or less; .25 if 9-12 grades with no diploma or equiv-
alency; .50 if 12 grades diploma or equivalency; .75 if some college; 1 if
college degree.
Class: "Do you consider yourself a member of the upper class, middle class,
working class, lower class, or are you not sure?" Coded 1 if upper class; .66 if
middle class; .33 if working class; 0 if anything else.
Employed: "Do you currently work outside of the home?" Coded 1 if yes; 0 if
no.
Targeted States: Coded 1 if Connecticut, Georgia, Kansas, Louisiana, Mon-
tana, Nebraska, New Jersey, North Dakota, Oklahoma, Rhode Island, Ten-
nessee; all other states coded 0.
Times-Mirror Center for the People and the Press New Political
Landscape Survey
Sample: National sample of 3,800 (voting-age adults)
Field Date: July 12-25, 1994
Communicated with Congress Ever: "Have you ever called, sent a letter, or
faxed a letter to your congressional representative or Senator to express your
opinion on an issue?" Coded 1 if yes; 0 if no.
Communicated with Congress In Past Year:1 Question wording. "Have you done
so in just the past 12 months?" Coded 1 if yes; 0 if no.
1
The following is the Times-Mirror Center's methodological note on measuring participa-
tion: "The question asked of respondents read as follows: People express their opinions
about politics and current events in a number of ways besides voting. I'm going to read you a
list of some of these ways. Please just tell me if you have or have not done each at some point
in the last 12 months.
Sources and Coding for Survey Data 135
Sex: By observation. Coded 1 if male; 0 if female.
Race: "What is your race? Are you African-American, white, Asian, or some
other?"
White: Coded 1 if white; 0 if anything else.
Marital Status: "Are you married, divorced, separated, widowed, or have
you never been married?" Coded 1 if married; 0 if anything else.
Party Identification: "In politics today, do you consider yourself a Repub-
lican, a Democrat, or independent? (If Republican or Democrat) "Do you
consider yourself a strong (Republican/Democrat) or not a strong (Repub-
lican/Democrat)?" (If independent, other, or no preference) "As of today do
you lean more to the Republican Party or the Democratic party?"
thing else.
else.
Party Intensity: Coded 1 if very strong; .5 if not very strong; 0 if in-
dependent.
Union Household: "Are you or is your (husband/wife) a member of a labor
union?" Coded 1 if yes; 0 if no.
Age: "What is your age?"
Under 35: Coded 1 if under 35; 0 if anything else.
Over 65: Coded 1 if over 65; 0 if anything else.
Education: "What is the last grade or class that you completed in school?"
Coded 0 if 8 grades or less; .25 if 9-12 grades with no diploma or equiv-
alency; .50 if 12 grades with diploma or equivalency; .75 if some college; 1 if
college degree.
"While respondents were repeatedly reminded to think about whether they had done each
c
in the last 12 months/ the level of participation reported was extremely high for the time
frame asked about. Prior research also has shown that many survey respondents do not have
accurate recall about when they had actually engaged in a prior behavior.
"Monitoring of interviews also indicated that some respondents were probably likely to err
on the side of including some activity (giving themselves credit for something) they had done,
even if it was not in the last 12 months. In order to test this theory, and remove the error that
results from the social desirability component — where those respondents who had engaged in
an activity, but not in the last year said 'yes' to the initial question - the Times-Mirror Center
conducted a second survey of participation.
"In the second survey respondents were first asked if they had ever done each of the
activities. Those saying they had done any activity were then asked to think 'only about the
last 12 months, that is since June of 1992,' and to tell interviewers if they had done each
activity in that specific time frame. With the social desirability component removed by
respondents being able to say they had 'ever' done an activity, the resulting '12 month'
figures are significantly lower. We feel the second measure is a much purer and more accurate
reading of what people have actually done in the last year."
136 Appendix A
Homeowner: "Do you own or rent your own home?" 1 if own; 0 if anything
else.
Talk Radio: "How often, if ever, do you listen to radio shows that invite
listeners to call in to discuss current events, public issues, and politics?"
Coded 1 if regularly; .66 if sometimes; .33 if rarely; 0 if never.
Rush Limbaugh: "I'd like to know how often you watch or listen to certain
TV and radio programs. As I read each, tell me if you watch or listen
regularly, sometimes, hardly ever, or never?" Coded 1 if regularly; .66 if
sometimes; .33 if hardly ever; 0 if never.
Efficacy: "Most elected officials care what people like me think." Coded 1 if
strongly agree; .75 if not strongly agree. "Most elected officials don't care
what people like me think." Coded 0 if strongly agree; .25 if not so strongly
agree.
Church Attendance: "Do you go to church, synagogue, or some place of
worship?" Coded 1 if every week; .75 if almost every week; .5 if once or twice
a month; .25 if a few times a year; 0 if never.
Political Information: Respondents were asked the following three informa-
tion questions. The information scale is based on the number of correct
answers: 1 for 3; .66 for 2; .33 for 1; 0 for O correct answers. "Now I'd like to
ask you about some things that have been in the news recently. Not everyone
will have heard about them all. Can you tell me the name of the current vice-
president of the United States? Do you happen to know which political party
has a majority in the U.S. House of Representatives? Can you tell me the
name of the President of Russia?"
Targeted Congressional District: Coded 1 if Slattery, KS-2; Cooper, TN-4;
Boucher, VA-9; Lehman, CA-19; Schenk, CA-49; Marolies-Mezvinsky,
PA-13; Lambert, AR-1; Tauzin, LA-3; Klug, WI-2; Upton, MI-6; Hall,
TX-4; Hoagland, NE-2; Brewster, OK-3; Payne, VA-5; Jefferson, LA-2;
Neal, MA-2; Kennelly, CT-1, Johnson, CT-6; Levin, MI-12. Otherwise
coded 0.
American National Election Study Survey

Sample: 1978 (2,304), 1980 (1,614), 1986 (2,176), 1990 (1,980), 1994
(1,795)
These data are taken from the American National Election Studies Cumula-
tive Data File, 1952-1992, and the 1994 Post Election Survey made avail-
able by the Inter-university Consortium for Political and Social Research.
Communicated with Congress: "Have you (or anyone in your family living
here) ever contacted Representative [name] or anyone in his or her office?
Sources and Coding for Survey Data 137
Was it to express an opinion, seek information, or to seek help on a problem
you had?" Coded 1 if yes; 0 if no.
Party Identification: "Generally speaking, do you usually think of yourself as
a Republican, a Democrat, an independent, or what?" (If Republican or
Democrat) "Would you call yourself a strong (Republican/Democrat) or not
very strong (Republican/Democrat)?" (If independent, other, or no prefer-
ence) "Do you think of yourself as closer to the Republican or Democratic
party?"
thing else.
else.
Appendix B
Interest Group Sampling Frame
Roster of Groups in Sampling Frame

Ad Hoc Coalition of Business for Health Care
Reform Trade Association
AFL-CIO Union
AETNA Corporation
AFSCME Union
Airline Pilots International Union
Alliance Pharmaceuticals Corporation
Allied Signal Corporation
American Association of Retired Persons Citizens' Group
American Automobile Manufacturers
Association Trade Association
American Bankers Association Trade Association
American Civil Liberties Union Citizens' Group
American College of Emergency Physicians Trade Association
American College of Nurse Midwives Trade Association
American College of Physicians Trade Association
American College of Surgeons Trade Association
American Conservative Union Citizens' Group
American Council of Life Insurance Trade Association
American Crop Protection Association Trade Association
American Dairy Producers Trade Association
American Dental Association Trade Association
American Federation of Teachers Union
American Farm Bureau Trade Association
American Hospital Association Trade Association
American Insurance Association Trade Association
American Legion Citizens' Group
American Life League Citizens' Group
138
Interest Group Sampling Frame 139
American Medical Association Trade Association

American Nurses Association Trade Association
American Petroleum Institute Trade Association
American Postal Workers Union Union
American Trial Lawyers Association Trade Association
American Trucking Association Trade Association
Americans for Democratic Action Citizens' Group
Anheuser-Busch Corporation
Anti-Defamation League Citizens' Group
AT&T Corporation
Bankcardholders of America Citizens' Group
Beer Institute Trade Association
Bell Atlantic Corporation
Bell South Corporation
Blue Cross/Blue Shield Corporation
Boeing Corporation
Brotherhood of Locomotive Engineers Union
Brown and Williamson Corporation
Burger King Corporation
Burroughs Welcome Corporation
Cellular Telecommunications Industry
Association Trade Association
Chemical Manufacturers Association Trade Association
Chem-Nuclear Systems Trade Association
Chevron Corporation
Children's Defense Fund Citizens' Group
Chiropractors Trade Association
Christian Coalition Citizens' Group
Chrysler Corporation
Ciba Pharmaceuticals Corporation
CIGNA Corporation
Citizen Action Citizens' Group
Citizen's Committee for the Right to Keep and
Bear Arms Citizens' Group
Citizens for a Sound Economy Citizens' Group
Clean Water Action Project Citizens' Group
Coalition to Stop Gun Violence Citizens' Group
Common Cause Citizens' Group
Communications Workers of America Union
Coors Corporation
COPE Union
Democratic National Committee Citizens' Group
140 Appendix B
Distilled Spirits Council Trade Association
Dow Chemical Corporation
DuPont Corporation
Eagle Forum Citizens' Group
Eastman Kodak Corporation
Edison Electric Institute Trade Association
Eli Lilly Corporation
Environmental Defense Fund Citizens' Group
Exxon Corporation
Families USA Citizens' Group
Family Research Council Citizens' Group
Federation for American Immigration Reform Citizens' Group
Federation of American Health Systems Trade Association
Fertilizer Institute Trade Association
Focus on the Family Citizens' Group
Food and Commercial Workers Union Union
Ford Corporation
General Electric Corporation
General Motors Corporation
Glaxo Corporation
Golden Rule Insurance Company Corporation
Grocery Manufacturers of America Trade Association
Group Health Association of American Trade Association
Gun Owners of America Citizens' Group
Handgun Control Citizens' Group
Health Care Reform Project Citizens' Group
Health Insurance Association of America Trade Association
Healthcare Leadership Council Trade Association
Household International Corporation
Human Rights Campaign Fund Citizens' Group
IBM Corporation
Independent Insurance Agents of America Trade Association
International Brotherhood of Electrical Workers Union
International Ladies Garment Workers Union Union
ITT Hartford Corporation
Izaak Walton League Citizens' Group
League of Conservation Voters Citizens' Group
McDonnell Douglas Corporation
MCI Corporation
Merck Corporation
Metropolitan Life Corporation
Mobil Oil Corporation
Mortgage Bankers Association of America Trade Association

Mutual of Omaha Corporation
NAACP Citizens' Group
National Gay and Lesbian Task Force Citizens' Group
National Abortion Rights Action League Citizens' Group
National Association of Broadcasters Trade Association
National Association of Health Underwriters Trade Association
National Association of Home Builders Trade Association
National Association of Letter Carriers Union
National Association of Manufacturers Trade Association
National Association of Wholesalers and
Distributors Trade Association
National Bankers Association Trade Association
National Coalition to Ban Handguns Citizens' Group
National Committee to Preserve Social Security
and Medicare Citizens' Group
National Conference of Catholic Bishops Citizens' Group
National Corn Growers Association Trade Association
National Cotton Council Trade Association
National Council of Senior Citizens Citizens' Group
National Education Association Union
National Farmers Association Citizens' Group
National Federation of Independent Businesses Trade Association
National Grange Union
National Grocers Association Trade Association
National Health and Human Services Employers
Union Union
National Milk Producers Federation Trade Association
National Organization for Women Citizens' Group
Natural Resources Defense Council Citizens' Group
National Restaurant Association Trade Association
National Retail Federation Trade Association
National Right to Life Committee Citizens' Group
National Right to Work Committee Citizens' Group
National Soft Drink Association Trade Association
National Taxpayers Union Citizens' Group
Nationwide Insurance Companies Corporation
Newspaper Guild Union
Nuclear Energy Institute Trade Association
NYNEX Corporation
Pacific Gas and Electric Corporation
People for the American Way Citizens' Group
142 Appendix B
Pfizer Corporation
Pharmaceutical Manufacturers Association Trade Association
Phillip Morris Corporation
Planned Parenthood Citizens' Group
Procter and Gamble Corporation
Public Citizen Citizens' Group
R. J. Reynolds Corporation
Rainforest Action Network Citizens' Group
Republican National Committee Citizens' Group
Schering Plough Corporation
Service Employees International Union Union
Sheetmetal Workers Association Union
Shell Oil Corporation
Sierra Club Citizens' Group
Smith Kline Beecham Corporation
Smokeless Tobacco Council Trade Association
Sprint Corporation
Teamsters Union
Texaco Corporation
Texas Medical Association Trade Association
Tobacco Institute Trade Association
Traveler's Insurance Corporation
Union Pacific Corporation
United Airlines Corporation
United Auto Workers Union
United Brotherhood of Carpenters and Joiners Union
United Food and Commercial Workers Union
United Mineworkers Union
United Seniors Association Citizens' Group
United Steelworkers of America Union
Upjohn Corporation
US Tobacco Corporation
U.S. Chamber of Commerce Trade Association
U.S. English Citizens' Group
U.S. Feed Grains Council Trade Association
U.S. PIRG Citizens' Group
US West Corporation
Veterans of Foreign Wars Citizens' Group
Voters for Choice Citizens' Group
Warner Lambert Corporation
Wine and Spirits Wholesalers of America Trade Association
World Wildlife Fund Citizens' Group
Xerox Corporation
Interview Schedule
1. Since President Clinton took office, what are some issues for which you
used grass roots mobilization to encourage constituents to contact their
representative or senator directly?
2. Why did you choose to use a grass roots strategy in the case of ?
3. Were there specific districts or states that you targeted in your lobbying
around ?
4. In working on , how did you choose which districts or states to
target?
5. Once you chose which members and districts needed to be targeted for
, how did you decide whom to mobilize within the district?
6. How did you develop specific mobilization messages for ?
7. What methods did you use to mobilize?
Appendix C
Chronology of Health Care Reform Legislation
Before examining the strategic and tactical choices of grass roots lobbyists, it
is important to understand the basic policy issues, political environment,
and chronology of events in the battle over the Clinton plan and health care
reform. (For more comprehensive accounts, see Johnson and Broder 1996;
Congressional Quarterly Almanac 1994; Skocpol 1996; and Clymer et al.
1994.)
The 1,364-page Clinton plan was ambitious. It promised guaranteed
coverage for all Americans by 1998. The funding mechanism was an em-
ployer mandate that would have required employers to pay 80 percent of
their workers' health insurance costs. The plan included spending caps on
health insurance premiums, a requirement that drug companies pay rebates
to the Medicare program, and a national health board responsible for re-
viewing prices for new drugs and establishing global budgets to limit the
nation's total spending on health care. In addition, the Clinton plan re-
quired Americans to purchase their health insurance through large purchas-
ing groups dubbed alliances. The plan also included a seventy-five-cents-
a-pack increase in the cigarette tax and a levy on large employers that did not
join the health alliances (Pear 1996).
As the Clinton plan was referred to committee there were three other
major competing plans. To the left of the Clinton plan was the single-payer
proposal introduced by Representative Jim McDermott of Washington
State. This plan was patterned after the Canadian system and would have
replaced private insurance companies with the federal government. With the
single-payer plan the federal government would have collected all premiums
and paid all health care providers. To the right of the Clinton plan were
separate bills proposed by Representative Jim Cooper of Tennessee and
Senator John Chafee of Rhode Island. Neither of these plans included an
employer mandate and both aimed to reorganize the health care market
rather than introduce new government legislation {Congressional Quarterly
Almanac 1994, 324).
144
Chronology of Health Care Reform 145
The following time line, taken from the Congressional Quarterly Almanac,
charts the chronology of the major events in the battle over health care
reform (1994, 321).
January 25, 1993: President Clinton appoints Hillary Rodham Clinton to

run the Health Care Reform Task Force.
March 3,1993: Representative Jim McDermott (D-Washington) introduces
a bill to set up a Canadian style single-payer system.
September 22, 1993: President Clinton's speech to Congress introduces the
administration's Health Care Reform Bill.
October 6, 1993: Representative Jim Cooper (D-Tennessee) introduces his
version of health care reform (Clinton-lite). It does not contain employer
mandates. Similar measure is also introduced in Senate by John Breaux of
Louisiana.
November 20,1993: Clinton's health care bill is officially introduced. House
leaders refer it to the Education and Labor, Energy and Commerce, and
Ways and Means committees.
November 22,1993: Senator John Chafee (R-Rhode Island) introduces a bill
requiring that all Americans have health insurance by 2005 (an individual
mandate).
January 25, 1994: In his State of the Union Address, President Clinton
threatens to veto any bill without universal coverage.
May 31, 1994: Representative Dan Rostenkowski of Illinois is indicted and
resigns as Chairman of House Ways and Means.
June 9, 1994: Senate Labor and Human Resources votes 11-6 along mainly
partisan lines to approve a bill modeled on the Clinton plan.
June 23,1994: House Education and Labor votes out an expanded version of
the Clinton bill. To assuage more liberal members, the committee also
sends to the floor without recommendation a single-payer plan.
June 28, 1994: House Energy and Commerce committee chairman John
Dingell (D-Michigan) notifies leadership that his panel cannot agree on
healthcare.
June 30,1994: House Ways and Means Committee approves a Clinton style
bill 20-18. No votes included all 14 Republicans and 4 Democrats.
July 2,1994: Senate Finance reports a bill with no employer mandates.
July 29, 1994: A House leadership plan based on Ways and Means bill is
unveiled. Employer mandate stays in and Medicare Part C is proposed.
August 2,1994: The Mitchell plan is unveiled in Senate.
August 11,1994: A rule to bring crime bill to a floor vote in House fails.
August 26,1994: Congressional leaders announce that no health care reform
bill will be brought up in session.
September 26, 1994: Last-chance compromise bill by mainstream group is
withdrawn.
Bibliography
Abramson, Paul. 1983. Political Attitudes in America. San Francisco: W. H. Freeman.

Abramson, Paul, and John Aldrich. 1982. "The Decline in Electoral Participation in America."
American Political Science Review 76:502—521.
Ainsworth, Scott, and Itai Sened. 1993. "The Role of Lobbyists: Entrepreneurs with Two
Audiences." American Journal of Political Science 37: 834-866.
Aldrich, John. 1993. "Rational Choice and Turnout." American Journal of Political Science
37:246-278.
Arnold, R. Douglas. 1990. The Logic of Congressional Action. New Haven, Conn.: Yale Univer-
sity Press.
Austen-Smith, David. 1993. "Information and Influence: Lobbying for Votes and Agendas."
American Journal of Political Science 37:799-833.
Austen-Smith, David, and John Wright. 1994. "Counteractive Lobbying." American Journal of
Political Science 38:25-44.
Austen-Smith, David, and John Wright. 1996. "Theory and Evidence for Counteractive Lob-
bying." American Journal of Political Science 40:543-564.
Barone, Michael, and Grant Ujifusa. 1993. Almanac of American Politics, 1994. Washington,
D.C.: National Journal Press.
Barone, Michael, and Grant Ujifusa. 1995. Almanac of American Politics, 1996. Washington,
D.C.: National Journal Press.
Bauer, Raymond, Ithiel de Sola Pool, and Lewis Dexter. 1964. American Business and Public
Policy. New York: Atherton Press.
Baumgartner, Frank, and Beth Leech. 1996a. "The Multiple Ambiguities of Counteractive
Lobbying." American Journal of Political Science 40:521-542.
Baumgartner, Frank, and Beth Leech. 1996b. "Good Theory Deserves Good Data." American
Journal of Political Science 40:565-572.
Bennett, Earl. 1986. Apathy in America, 1960-1984: Causes and Consequences of Citizen Political
Indifference. Dobbs Ferry, N.Y.: Transnational Publishers.
Berry, Jeffrery. 1977. Lobbying for the People. Princeton, N.J.: Princeton University Press.
Berry, Jeffrey. 1989. The Interest Group Society. Glenview, 111.: Scott Foresman.
Brady, Henry, and Cara Buckley. 1995. "Health Care Reform in the 103rd Congress: A
Predictable Failure." Journal of Health Politics, Policy, and Law 20 (Summer): 403-410.
Brinkley, Joel. 1993a. "Lobbying Rules of the 1990's Show the Most Vulnerable." New York
Times, June 16, p. Al.
Brinkley, Joel. 1993b. Cultivating the Grass Roots to Reap Legislative Benefits. New York
Times, November 1, p. Al.
147
148 Bibliography
Browne, William. 1988. Private Interests, Public Policy, and American Agriculture. Lawrence:
University Press of Kansas.
Browne, William. 1990. "Organized Interests and Their Issue Niches: A Search for Pluralism
in the Policy Domain." Journal of Politics 52:477-509.
Browne, William. 1995. Cultivating Congress: Constituents, Issues, and Interests in Agricultural
Policymaking. Lawrence: University Press of Kansas.
Browning, Graeme. 1994. "Zapping the Capitol." National Journal, October 22, pp. 2446-
2450.
Campbell, Angus, Philip Converse, Warren Miller, and Donald Stokes. 1960. The American
Voter. New York: Wiley.
Carlson, Peter. 1990. "The Image Makers." Washington Post Magazine. February 11, p. W12.
Carney, Eliza Newlin. 1997. "Stealth Bombers." National Journal, August 16, pp. 1640-1645.
Chandler, Clay. 1995. "Bellowing for a Balanced Budget." Washington Post, September 7, p.
B12.
Chong, Dennis. 1991. Collective Action and the Civil Rights Movement. Chicago: University of
Chicago Press.
Cigler, Allan, and Burdett A. Loomis. 1995. Interest Group Politics. Washington, D.C.: Con-
gressional Quarterly Press.
Clymer, Adam. 1993. "Doctors Soften Criticism of Health Care Plan." New York Times,
October 17, p. A14.
Clymer, Adam, Robert Pear, and Robin Toner. 1994. "For Health Care Time Was a Killer."
New York Times, August 29, p. Al.
Cohen, Richard. 1992. Washington at Work. New York: Macmillan.
Congressional Quarterly Almanac. 1994. Washington, D.C.: Congresional Quarterly Press.
Conway, Margaret. 1991. Political Participation in the United States. Washington, D.C.: Congre-
tional Quarterly Press.
Cutright, Phillips. 1963. "Measuring the Impact of Local Party Activity on the General Elec-
tion Vote." Public Opinion Quarterly 27:372-386.
Dexter, Lewis. 1970. "What do Congressmen Hear: The Mail." Public Opinion Quarterly 20
(Spring): 523-584.
Dowd, Ann. 1993. "How to Get Things Done in Washington." Fortune, August, 9, p. 60.
Downs, Anthony. 1957. An Economic Theory of Democracy. New York: Harper Row.
Drew, Christopher, and Michael Hackett. 1992. "More and More Lobbyists Call Shots in
DC." Chicago Tribune, December 6, p. Al.
Duncan, David Ewing. 1994. "The Triumph of Harry and Louise: How the Mighty Insurance
Lobby Skunked Clinton and Demolished His Health-Care Plan." Los Angeles Times Maga-
zine, September 11, p. 28.
Fenno, Richard. 1978. Home Style: House Members in Their Districts. Boston: Little Brown.
Fenno, Richard. 1973. Committees in Congress. Boston: Little Brown.
Fowler, Linda, and Robert McClure. 1989. Political Ambition: Who Decides to Run for Congress?
New Haven: Yale University Press.
Fowler, Linda, and Ronald Shaiko. 1987. "The Grass Roots Connection: Environmental
Activists and Senate Roll Calls." American Journal of Political Science 31:484-510.
Frantzich, Stephen. 1986. Write Your Congressman. New York: Praeger.
Fritsch, Jane. 1995. "Special Pleaders: The Lobbyists' Art." New York Times, March 18, p. Al.
Fritsch, Jane. 1996. "Sometimes Lobbyists Strive to Keep Public in the Dark." New York Times,
March 18, p. Al.
Gant, Michael, and Norman Luttbeg. 1991. American Electoral Behavior. Itasca, 111.: F. E.
Peacock.
Goldstein, Kenneth. 1994. "Mobilization and Participation over Time." Paper presented at
Bibliography 149
the annual meeting of the Midwest Political Science Association, Chicago, April 14-16.
Goldstein, Kenneth. 1995. "Seeding the Grass Roots: Mobilization and Contacting Con-
gress." Paper presented at the annual meeting of the Midwest Political Science Associa-
tion, Chicago, April 6-8.
Gosnell, Harold. 1927. Getting out the Vote. Chicago: University of Chicago Press.
Greider, William. 1992. Who Will Tell the People. New York: Simon and Schuster.
Grenzke, Janet. 1989a. "Shopping in the Congressional Supermarket: The Currency Is Com-
plex." American Journal of Political Science 33:1-24.
Grenzke, Janet. 1989b. "Candidate Attributes and PAC Contributions." Western Political Sci-
ence Quarterly 42:245-264.
Gugliotta, Guy. 1994. "A Man Who Fertilizes the Grass Roots." Washington Post, August 23, p.
A10.
Hall, Richard. 1987. "Participation and Purpose in Committee Decision Making." American
Political Science Review 81:105-127.
Hall, Richard, and Wayman, Frank. 1990. "Buying Time: Moneyed Interests and the Mobili-
zation of Bias in Congressional Committees." American Political Science Review 84:797-
820.
Hansen, John Mark. 1991. Gaining Access: Congress and the Farm Lobby. Chicago: University of
Chicago Press.
Harris, John. 1995. "Law Aspiring to Shed Light on Lobbyists Leaves Some Gray Areas."
Washington Post, December 20, p. A4.
Hayes, Michael. 1981. Lobbyists and Legislators: A Theory of Political Markets. New Brunswick,
N.J.: Rutgers University Press.
Headen, Susan. 1994. "The Little Lobby That Could." U.S. News and World Report, Septem-
ber 12, p. 45.
Healthline. 1993-1994. Alexandria, Va.: American Political Network.
Herring, Pendleton. 1929. Group Representation before Congress. Baltimore: Johns Hopkins
University Press.
Hosenball, Mark. 1989. "Letters Sealed with Fear: Direct-Mail Moguls and the Catastrophic-
Care Bill." Washington Post, October 22, p. Cl.
Hotline. 1995. Alexandria, Va.: American Political Network. August 23.
Hotline. 1996. Alexandria, Va.: American Political Network. July 12.
Houston, Paul. 1993. "Phone Frenzy in the Capitol - Special Interests Are Using Sophisti-
cated Electronic Networks to Generate an Astonishing Number of Calls to Congress." Los
Angeles Times, May 16, p. Al.
Iyengar, Shanto, and Donald Kinder. 1987. News that Matters. Chicago: University of Chicago
Press.
Jacobson, Gary. 1992. The Politics of Congressional Elections. New York: Harper Collins
Publishers.
Jacobson, Gary, and Samuel Kernell. 1981. Strategy and Choice in Congressional Elections. New
Haven: Yale University Press.
Jacobson, Gary, and Samuel Kernell. 1990. "National Forces in the 1986 House Elections."
Legislative Studies Quarterly 15:72-85.
Jamieson, Kathleen Hall. 1994. "The Role of Advertising in the Health Care Reform Debate."
University of Pennsylvania press release, July 18.
Jennings, Kent, and Richard Niemi. 1974. The Political Character ofAdolescence: The Influence of
Families and Schools. Princeton, N.J.: Princeton University Press.
Johnson, Haynes, and David Broder. 1996. The System. Boston: Little Brown.
Katz, Daniel, and Samuel Eldersveld. 1961. "The Impact of Local Party Activity upon the
Electorate." Public Opinion Quarterly 25:1-24.
150 Bibliography
Keller, Bill. "Special-Interest Lobbyists Cultivate the 'Grass Roots' to Influence Capitol Hill."
Congressional Quarterly, September 12, pp. 1739-1742.
Kernell, Samuel. 1993. Going Public. Washington, D.C.: Congressional Quarterly Press.
Kingdon, John. 1989. Congressmen's Voting Decisions. Ann Arbor: University of Michigan Press.
Knight, Jerry. 1991. "A Night to Play Let's Make a Deal." Washington Post, November 28, p.
Al.
Kosterlitz, Julie. 1993. "Health Lobby Pushes Past the Beltway. National Journal, April 17, pp.
34-39.
Kramer, Gerald. 1971. "The Effects of Precinct Level Canvassing on Voting Behavior." Public
Opinion Quarterly 34:560-572.
Krasno, Jonathan, and Donald Green. 1988. "Preempting Quality Challengers in House Elec-
tions." Journal of Politics 50: 920-936.
Krehbiel, Keith. 1991. Information and Legislative Organization. Ann Arbor: University of
Michigan Press.
Leech, Beth. 1997. "Conflictual and Cooperative Lobbying Strategies." Paper presented at the
annual meeting of the American Political Science Association, Washington, D.C., August
28-31.
Levin, Carl. 1994. "Lobbying Disclosure Conference Report." Congressional Record-Senate,
October 3, p. SI3945.
Lewis, Neil. 1994. "Lobby for Small-Business Owners Puts Big Dent in Health Care Bill."
New York Times, July 5, p. Al.
Mann, Thomas, and Norman Ornstein. 1994. Vital Statistics on Congress, 1993-1994. Wash-
ington, D.C.: Congressional Quarterly Press.
Martin, Cathie Jo. 1995. "Stuck in Neutral: Big Business and the Politics of National Health
Care Reform." Journal of Health Politics, Policy, and Law 20 (Summer): 431-436.
Mathews, Donald. 1960. U.S. Senators and Their World. Chapel Hill: University of North
Carolina Press.
Mathews, Donald, and James Stimson. 1975. Yeas and Nays: Normal Decision-Making in the
United States House of Representatives. New York: James Wiley.
Mayhew, David. 1974. Congress: The Electoral Connection. New Haven: Yale University Press.
McCarthy, John and Mayer Zald. 1977. "Resource Mobilization and Social Movements."
American Journal of Sociology 82 (May): 1212-1241.
McDonald, John. 1994. "Health Insurance Money Flows to Lieberman." Hartford Courant,
June 9, p. Al.
Milbrath, Lester. 1963. The Washington Lobbyists. Westport, Conn.: Greenwood Press.
Milbrath, Lester, and M. L. Goel. 1977. Political Participation: How and Why Do People Get
Involved. Chicago: Rand McNally.
Mintz, John. 1994. "Some Firms Urging Workers to Oppose Health Care Bill." Washington
Post, August 20, p. A6.
Mitchell, Alison. 1995. "Clinton Applies '94 Lesson to Fight G.O.P. Budget." New York Times,
October 12, p. Al.
Mitchell, William, and Michael Munger. 1991. "Economic Models of Interest Groups: An
Introductory Survey." American Journal of Political Science 35:512-546.
Nagel, Jack. 1987. Participation. Englewood Cliffs, N.J.: Prentice-Hall.
Newsweek. 1991. "Watering the Grass Roots." May 6, p. 36.
Neumann, W. Russell. 1986. The Paradox of Mass Politics and Opinion in the American Electorate.
Cambridge, Mass.: Harvard University Press.
Odegard, Peter. 1928. Pressure Politics: The Story of the Anti-Saloon League. New York: Colum-
bia University Press.
Bibliography 151
Olson, Mancur. 1965. The Logic of Collective Action. Cambridge, Mass.: Harvard University
Press.
Ornstein, Norman, and Shirley Elder. 1978. Interest Groups, Lobbying and Policy-making. Wash-
ington, D.C.: Congressional Quarterly Press.
Pateman, Carol. 1970. Participation and Democratic Theory. Cambridge: Cambridge University
Press.
Petracca, Mark. 1992. The Politics of Interests: Interest Group Politics Transformed. Boulder, Colo.:
Westview Press.
Powell, Bingam. 1986. "American Turnout in Comparative Perspective." American Political
Science Review 80:17-44.
Riker, William, and Peter Ordeshook. 1968. "A Theory of the Calculus of Voting." American
Political Science Review 62 (June): 2 5 - 4 2 .
Romer, Thomas, and Howard Rosenthal. 1978. "Political Resource Allocation." Public Choice
33:27-45.
Rosenau, James. 1974. Citizenship between Elections: An Inquiry into the Mobilizable American.
New York: Free Press.
Rosenstone, Steven, and Mark Hansen. 1993. Mobilization, Participation, and Democracy in
America. New York: Macmillan Press.
Rothenberg, Lawrence. 1992. Linking Citizens to Government. Cambridge: Cambridge Univer-
sity Press.
Scarlett, Thomas. 1994. "Killing Health Care Reform." Campaigns and Elections 15 (October-
November): 34-37.
Schattschneider, E. E. 1960. The Semi-Sovereign People: A Realist's View of Democracy in Amer-
ica. New York: Holt, Rinehart and Winston.
Schlozman, Kay, and John Tierney. 1983. "More of the Same: Washington Pressure Group
Activity in a Decade of Change." Journal of Politics 45: 353-373.
Schlozman, Kay, and John Tierney. 1986. Organized Interests and American Democracy. New
York: Harper and Row.
Seelye, Katharine. 1994. "Hobbling of Lobbying Bill Shows Muscle Power of'Grass Roots'
Conservative Network." New York Times, October 7, p. A22.
Skocpol, Theda. 1996. Boomerang. New York: W W. Norton.
Smith, Richard. 1984. "Advocacy Interpretation and Influence in the United States Con-
gress." American Political Science Review 78:44-63.
Stone, Peter. 1993. "Green Green Grass." National Journal March 27, pp. 754-757.
Stone, Peter. 1994. "Health Care Reform's Price Fight." National Journal, August 13, pp.
1923-1924.
Taylor, Paul. 1983. "The Death of Withholding, or How the Bankers Won Big." Washington
Post, July 31, p. A12.
Thompson, Margaret. 1985. The Spider Web. Ithaca, N.Y.: Cornell University Press.
Tierney, John. 1992. "Organized Interests in the Nation's Capitol." In Mark Petracca, ed., The
Politics of Interests: Interest Group Politics Transformed, pp. 201-20. Boulder, Colo.: West-
view Press.
Toner, Robin. 1994. "Harry and Louise and a Guy Named Ben." New York Times September
23, p. A8.
Truman, David. 1951. The Governmental Process. New York: Alfred Knopf.
Tversky, Amos, and Daniel Kahneman. 1981. "The Framing Decisions and the Psychology of
Choice." Science 211:453-458.
Verba, Sidney, and Norman Nie. 1972. Participation in America: Political Democracy and Social
Equality. New York: Harper and Row.
152 Bibliography
Verba, Sidney, Kay Schlozman, and Henry Brady. 1995a. Voice and Equality. Cambridge,
Mass.: Harvard University Press.
Verba, Sidney, Kay Schlozman, and Henry Brady. 1995b. "Beyond SES: A Resource Model of
Political Participation." American Political Science Review 89 (June): 271-294.
Vogel, David. 1989. Fluctuating Fortunes. New York: Basic Books.
Walker, Jack. 1991. Mobilizing Interests in America. Ann Arbor: University of Michigan Press.
Weisskopf, Michael. 1993. "Lining up Allies in the Health Care Debate." Washington Post,
October 3, p. A4.
West, Darrell, Diane Heith, and Chris Goodwin. 1995. "Political Advertising and Health Care
Reform." Paper presented at the annual meeting of the Midwest Political Science Associa-
tion, Chicago, April 6-8.
Wielhouwer, Peter, and Brad Lockerbie. 1994. "Party Contacting and Political Participation."
American Journal of Political Science 38 (February): 211-229.
Wilson, James. 1973. Political Organization. New York: Basic Books.
Wittenberg, Ernest and Elisabeth Wittenberg. 1994. How to Win in Washington, Cambridge,
Mass.,: Blackwell.
Wolfinger, Raymond. 1963. "The Influence of Precinct Work on Voting Behavior." Public
Opinion Quarterly 27:387-398.
Wolfinger, Raymond, and Steven Rosenstone. 1980. Who Votes. New Haven: Yale University
Press.
Wolpe, Bruce. 1990. Lobbying Congress. Washington, D.C.: Congressional Quarterly Press.
Woodward, Bob. 1996. The Choice. New York: Simon and Schuster.
Wright, John. 1985. "PACs, Contributions, and Roll Call: An Organizational Perspective."
American Political Science Review 79:400-414.
Wright, John. 1990. "Contributions, Lobbying and Committee Voting in the U.S. House of
Representatives." American Political Science Review. 84:417-438.
1996. Interest Groups and Congress. Boston: Allyn and Bacon.
Zaller, John. 1992. The Nature and Origins of Mass Opinion. Cambridge: Cambridge University
Press.
Zeigler, Harmon. 1964. Interest Groups in American Society. Englewood Cliffs, N.J., Prentice-
Hall.
Ziegler, Harmon, and Wayne Peak. 1972. Interest Groups in American Society. Englewood Cliffs,
N.J.: Prentice-Hall.
Zorack, John. 1993. The Lobbying Handbook. Washington, D.C.: Professional Lobbying and
Consulting Center.
Index
ABC News, 75 Balanced Budget Amendment, 7, 54, 56

ABC News/Washington Post poll, 73 banking industry, 1, 3
abortion, 7, 20, 40, 54, 60 Barbour, Haley, 40
Abramson, Paul, 18 Barone, Michael, 80, 89, 90, 97, 101
advertising Battleground Poll, 6, 13, 107-117,
newspaper, 1 121
radio, 26, 87, 108 Baucus, Max, 101, 102, 103, 108
television, 26, 27, 31, 34, 40, 41, 70, 87, Bauer, Raymond, 23, 127
108 Baumgartner, Frank, 46
Aetna, 104 Bennet, Earl, 18, 127
AFL-CIO, 26, 70 Bentley Historical Library, 14
AIDS, 65 Berelson, 129
Ainsworth, Scott, 127 Berry, Jeffrey, 2, 10
Aldrich, John, 18, 19 "Big Three" automakers, 2, 65
allies, 34, 35, 51, 57, 60, 65, 69, 84, 88 Bliley, Thomas, 105
Almanac of American Politics (1994), 80 Bonior, David, 55
American Association of Retired Persons Bonner, Jack, 2, 3, 62, 65
(AARP), 68, 85, 125 Bonner and Associates, 2
American Bankers Association (ABA), 1, 3 Boren, David, 60, 101, 102, 103, 108
American Civil Liberties Union (ACLU), Boucher, Rick, 90, 91, 94, 118
130 Boy Scouts of America, 3, 65, 126
American Conservative Union, 79 Bradley, Bill, 101, 102, 103, 108
American Cotton Manufacturers Institute Brady, Henry, 4, 17-21, 105
(ACMI), 23 Breaux, John, 101-103, 108, 145
American Energy Alliance, 125 Brewster, Bill, 60, 97, 100, 118
American Federation of State, County, and Brinkley, Joel, 2, 41
Municipal Employees, 25 Broder, David, 73-75, 82, 88, 95, 97, 101,
American Federation of Teachers, 82 104, 144
Americans for Democratic Action (ADA), Browne, William, 10
79, 80, 90-93, 97-99, 101, 102 Browning, Graeme, 2
Andrews, Michael, 97, 98, 100 BTUtax, 125, 126
Animal Farm, 30 Buckley, Cara, 105
anti-environmentalists, 3 budget bill (or Clinton budget), 7, 27, 33,
Anti-Saloon League, 22 54, 56, 57, 67, 68, 72
Archer, Bill, 98 Bunning, Jim, 98
Arnold, R. Douglas, 31, 32, 33, 36, 38, 76 Bush, George, 68, 91, 97
assault weapon ban, 40
Austen-Smith, David, 46, 127 Camp, Dave, 98
Campaign finance, 7, 54
Bailey, Pam, 84 Campaigns and Elections magazine, 25
153
154 Index
Campbell, Angus, 18, 127 letters, 3, 14, 15, 16, 17, 21, 22, 23, 24,
Cardin, Benjamin, 98 62, 64, 65, 67, 73, 110, 113, 125
Carney, Eliza Newlin, 27 phone calls (to Congress), 3, 14, 15, 17,
Carter, Jimmy, 122 24, 26, 62, 64, 65, 67, 73, 110, 113,
Catastrophic Care legislation, 2 125
CBS News, 75, 88 spontaneous communications, 45, 50
Chafee, John, 101-103, 108, 144, 145 stimulated communications, 1, 2, 4, 7, 8,
challenger quality (or quality of chal- 10-13, 21, 22, 27, 28, 30, 31, 34, 35,
lengers), 31, 32, 34, 35, 46, 67 39, 41-46, 49, 52, 56-61, 63, 65-67,
Chong, Dennis, 64 70, 74-78, 86, 87, 88, 91, 94, 96, 103,
Christian Coalition, 67, 85 117, 128, 129, 130
Chrysler Corporation, 65 telegrams, 14, 15, 17, 23, 24, 73, 88, 125
Cigler, Allan, 23 constituent intensity, role in legislative deci-
Citizens Action, 26 sion making, 36-39, 41, 42, 45, 48, 55
Citizens for a Sound Economy, 72, 94 content of messages, 42, 43, 44
Citizens for Reform, 27 Contract with America, 57
Civic Voluntarism Model, 20, 21 Converse, Philip, 18, 127
civil rights movement, 64 Conway, Margaret, 18
Clean Air Act, 2, 3, 65 Cook Report, 79
Clinton, Bill, 27, 40, 41, 46, 54, 56, 57, 67, Cooper, Jim, 90, 91, 94, 108, 118, 144,
68, 70-83, 85, 86, 88-91, 94, 96, 97, 145
100, 101, 106, 107, 115, 120-122, COPE, 12
144, 145 corporations, 24-26, 40, 54, 63, 66, 75, 76
Clinton, Hillary Rodham, 72-74, 95, 145 Coyne, William, 98
Clymer, Adam, 73, 100, 104, 144 Crane, Philip, 98
CNBC, 88 crime bill, 7, 54, 56, 57, 60, 125, 127, 145
CNN, 88 Crossfire, 40
Cohen, Richard, 3 cue givers, 47, 48, 49, 51, 58, 59, 78, 83,
collective action problem, 19 90, 104, 106, 123
Congressional committees, 47, 59, 78, 79, Cutright, Phillips, 21
84
House Banking, Finance, and Urban Af- Danforth, John, 101, 102, 103
fairs Committee, 3 Daschle, Tom, 12, 102
House Democratic Caucus, 12 deficit reduction, 57
House Education and Labor, 78-83, Democratic Congressional Campaign Com-
107, 108, 145 mittee, 34
House Energy and Commerce, 78-82, Democratic House Caucus, 120
89-99, 100, 101, 107, 117-119, 126, Democratic Leadership Council, 100, 101
145 Democratic National Committee, 26, 68,
House Rules Committee, 83 69
House Ways and Means Committee, 1, democrats, 1, 2, 6, 11, 12, 27',31, 34, 47,
78-82, 89, 96-100, 107, 117-119, 56, 57, 60, 68, 74, 79-84, 88-91, 95-
126, 145 97, 101, 103, 104, 107, 110, 112,
Congressional Quarterly Almanac (1994), 83, 120-123, 127, 145
90, 101, 104, 144, 145 Dexter, Lewis, 23, 127
Congressional Quarterly Weekly Report, 8 Dingell, John, 79, 89-91, 95, 96, 105, 145
Congressional Research Service, 15, 16 direct activities, 5, 23, 28, 35, 109, 111,
Conrad, Kent, 102, 103, 108 128
constituent communications (or citizens' direct costs, 50, 51, 56, 58, 78, 91, 127,
communications), 5, 6, 14-18, 20, 24, 129
39, 49, 51, 55, 66, 78, 83, 89, 104, direct mail/mailings, 2, 31, 94, 108
107, 109-112, 114-116, 118-123, district-matching software, 61
125, 126 Dixon, David, 24
e-mail, 15, 73, 95, 125 Dole, Robert, 1, 26, 101, 102
faxes, 3, 14, 15, 17, 24, 73, 95, 110, 113, Dorgan, Byron, 103, 108
125 Dowd, Ann, 24, 41
Index 155
Downs, Anthony, 19 Gramm, Phil, 46
Dukakis, Michael, 70 Grandy, Fred, 98
Dulles Airport, 10 grass roots communication, 3 9 - 4 2 , 47, 51,
Duncan, David Ewing, 74 52, 56
DuPont, 86 grass roots lobbying/mobilization/campaigns
Durenberger, Dave, 101, 102, 103 (or issue advocacy campaigns), 3 - 8 ,
11-13, 22-25, 27, 28, 30, 31, 34, 4 0 -
Eastman Kodak, 86 43, 45, 46, 53-57, 59-65, 67-69, 71,
Eldersveld, Samuel, 21 73, 74, 76-78, 83, 85, 91, 96, 100,
electoral (and direct electoral) objective, 4, 101, 103-109, 111, 115, 117, 123,
6, 12, 26, 28, 33, 39, 41, 4 6 - 4 9 , 5 3 - 125, 126, 128, 144
56, 58, 67-71, 76, 77, 91, 97, 126, grass tops, see key contacts
127 Grassley, Charles, 102
electoral connection, 28, 39, 45 Green, Donald, 35
environmental regulations, 69 Greenberg, Stanley, 57
environmentalists, 24-26, 55, 57, 58 Guardian Advisory Council (GAC), 87,
Exon, James, 104, 108 95
express-advocacy, 27 Gudermuth, Lori, 6
Gugliotta, Guy, 2
Far is, Jack, 95
Federal Elections Commission, 86 Hall, Ralph, 90, 91, 118
Fenno, Richard, 11, 36, 44 Hall, Richard, 42, 59, 94, 127
field staff, 31 Hancock, Mel, 98
filibuster, 101, 104 Hansen, John Mark, 10
Flaherty, Peter, 27 Hansen, Mark, 4, 14, 19, 21, 22, 50, 115
Foley, Tom, 12, 73, 83 "Harry and Louise" Campaign, 74, 77, 88,
Food and Drug Administration, 65 96,97
Ford, Harold, 98 Hart, Philip, 14
Ford, William, 79, 81 Hatch, Orrin, 102
Ford Motor Company, 64, 65 Hatfield, Matt, 103
Fortune 500 Companies, 25 Hayes, Michael, 127
Fowler, Linda, 34 Hayes, Patrick, 70
Framing Hayes, Ron, 70
issues, 39, 40, 41, 43, 45, 77, 86 Headen, Susan, 74
messages, 78 Headline News, 88
roll call votes, 39 health care reform, 7, 8, 13, 27, 46, 53, 54,
free riders, 19 57, 71, 72-111, 113, 115, 117, 120,
Freeport McMoran, 12 121, 123-126, 144, 145
Furse, Elizabeth, 12 Health Care Reform Bill, 145
Health Care Reform Task Force, 145
Gallup Organization, 73, 77 Health Insurance Association of America
Gant, Michael, 20 (HIAA), 73, 74, 84, 88, 96, 97
Gates, Bill, 43 Healthcare Leadership Council, 84
General Motors, 65 Healthline, 8
Gephardt, Dick, 12, 55, 73, 83 Heith, Diane, 77, 84, 88
get-out-the-vote (GOTV), 49, 90 Herger, Wally, 98
Gibbons, Sam, 97, 98 Herring, Pendleton, 22
Gingrich, Newt, 26, 40, 57, 120 Hoagland, Peter, 97, 98, 100, 118
Goddard, Ben, 87, 88 Hotline, 8, 26
Goeas, Ed, 6 Houghton, Amo, 98
Goel, M. L., 18 Human Rights Campaign Fund, 12
going public strategy, 3, 4, 6, 24 Hyde Amendment, 60
Goldstein, Kenneth, 19, 49, 115
Goodwin, Chris, 77, 84, 88 IBM, 86
Gosnell, Harold, 21 incentives/benefits to participate, 12, 16,
Gradison, Bill, 84 18, 19, 27
156 Index
incumbency, 31, 32, 33, 34, 35, 36, 46, 67, Levin, Carl, 14
128 Levin, Sander, 12, 99, 100, 118
indirect (lobbying) tactics, 24, 32, 34, 35, Lewis, John, 99
109 Lewis, Neil, 73, 74
individual pluralism, 24 liability laws, 65
information distribution, 39, 58, 65, 67, 76, Lieberman, Joseph, 103, 104, 108
83, 126, 129 Limbaugh, Rush, 88, 118-120
inside strategies, 3, 46, 60, 74, 104 lobbying reform, 7, 13, 54, 129
institutional pluralism, 24 Lockerbie, Brad, 49
internal group pressure, 64 Loomis, Burdett A., 23
Isakowitz, Mark, 95 Luttbeg, Norman, 20
Jackson, Brooks, 88 Magaziner, Ira, 73, 90, 96
Jacobs, Andy, Jr., 98 mainstream coalition, see "rump group"
Jacobson, Gary, 31, 34 Mann, Thomas, 16
Jamieson, Kathleen Hall, 77 Margolies-Mezvinsky, Marjorie, 90, 91, 94,
Jefferson, William, 98, 100, 118 105,118
Johnson, Haynes, 73, 74, 82, 95, 97, 101, Martin, Cathie Jo, 73
104, 144 Mathews, Donald, 49, 127
Johnson, Nancy, 97, 99, 100, 118 Matsui, Robert, 99
Johnston, J. Bennett, Jr., 103, 108 Mayhew, David, 36
McClure, Robert, 34
Kahneman, Daniel, 50 McCrery, Jim, 99
Katz, Daniel, 21 McCurdy, Dave, 60
Kennedy, Ted, 27, 46, 81-83 McDermott, Jim, 99, 100, 144, 145
Kennelly, Barbara, 99, 100, 118 McDonald, John, 104
Kernell, Samuel, 3, 24, 34 McNulty, Michael, 99
Kerrey, Bob, 103, 104, 108 meat inspection procedures, 7, 54, 57
key committee members, 3, 15, 47, 48, 51, Medicaid, 82
58-60, 72, 73, 78, 84, 85, 89, 95, 96, Medicare, 2, 7, 26, 33, 54, 57, 66, 68, 69,
97, 104, 106, 107, 115 70, 82, 125, 127, 144, 145
key contacts (or "grass tops"), 61-64, 70, Mellman-Lazarus-Lake consulting firm, 11,
87, 88, 106, 123 107
Kingdon, John, 10, 32, 36, 38, 45, 47, 52 message effectiveness, 71
Klug, Scott, 91, 94, 118 midnight basketball, 40, 56
Knight, Jerry, 4 Milbrath, Lester, 18, 127
Kopetski, Mike, 99 Miller, Warren, 18, 127
Kosterlitz, Julie, 88 Mintz John, 24, 86
Kramer, Gerald, 21 Mitchell plan, 145
Kramer, Tony, 25 Mitchell, George, 73, 82, 83, 102
Krasno, Jonathan, 35 Mobil, 86
Krehbiel, Keith, 83 mobilization, 3, 6, 8, 11, 12, 20, 21, 22, 25,
28, 30, 33-35, 41-43, 45-47, 49-52,
Lake, Celinda, 6 54-56, 58, 62-68, 72-75, 78, 81-85,
Lake Research, 6 87-91, 96, 104-106, 108-110, 112-
Lambert, Blanche, 90, 91, 94, 118 117, 123-125, 127, 128, 130, 131
LaRocco, Larry, 12 Mobilization, Participation, and Democracy in
Lautenberg, Frank, 103, 108 America, 14
League of Conservation Voters, 12, 25, 26 mobilization weight in strategic calculations,
Leech, Beth, 8, 46 44,45, 51
Legislate on-line service, 97, 101 Motley, John, 72
Legislative Demographic Services, 61 motor-voter legislation, 103
legislative objectives, 5, 12, 23, 40, 43, 45- Moynihan, Daniel Patrick, 81, 82, 101,
47, 51, 54-56, 58-61, 64, 67, 71, 73, 102, 103
75, 77, 89, 103, 106, 126
Lehman, Richard, 90, 91, 94, 95, 118 NAFTA, 7, 54, 55, 60, 63, 64, 85, 110,
Leo Burnett and Company, 2 126
Index 157
Nagel,Jack, 21 pork barrel legislation, 40, 56
National Committee to Preserve Social Se- Powell, Bingam, 21
curity and Medicare, 2 Price, David, 12
National Education Association, 82 priming, 40, 56
National Election Studies, 6, 16, 17, 21, Project 95, 25
22, 122, 123 Pryor, David, 102
National Federation of Independent Busi- Public Affairs Council, 8, 25
ness (NFIB), 72, 74, 87, 88, 90-99, Public Opinion Strategies, 6, 73
101, 102, 125
National Journal, 8, 24, 25, 70, 79 Rangel, Charles, 99
National Restaurant Association, 40, 41, 45 rational choice model, 19, 20, 44, 49
National Rifle Association, 40, 56, 120, Reagan, Ronald, 1, 122
125, 130 Reciprocal Trade Act, 23
Neal, Richard, 99, 100, 118 Republican National Committee, 27, 40,
Neumann, W. Russell, 18, 127 61, 120
New York Times, 7, 24, 25 Republicans, 1, 6, 26, 27, 57, 60, 68, 70,
Newsweek magazine, 2, 3 74, 79, 88, 90, 100, 101, 103, 107,
Nexis, 25 110, 112, 119-123, 127, 145
Nie, Norman, 18 representation ratios, 115, 116
Nunn, Sam, 47, 103, 104, 108 Riegle, Donald, Jr., 12
Nuttle, Marc, 88 Riker, William, 19
Roberts, Cokie, 75
Odegard, Peter, 22 Roberts, Steve, 75
Olson, Mancur, 19 Robertson, Pat, 88
Ordeshook, Peter, 19 Rockefeller, John, 102
organizational hurdles/resources, 11, 84, 85, Rockwell, Norman, 85
89 roll call vote, 5, 32, 39
organizational maintenance objectives, 55, Romer, Thomas, 47
75 Roosevelt, James, 2
organized labor, 25, 26, 70, 76 Roper Center Survey, 17, 21, 22
Ornstein, Norman, 16 Rosenstone, Steven, 4, 14, 18, 19, 21, 22,
Orwell, George, 30 50, 115
outside lobbying/outside tactics {see also Rosenthal, Howard, 47
grass roots lobbying), 3, 4, 5, 24, 28, Rostenkowski, Dan, 1, 79, 96, 97, 99, 145
38, 59, 60, 76, 77 Rotary Club, 45
Roth, Bill, 101, 102
Packwood, Bob, 101, 102 Rothenberg, Lawrence, 32, 52, 127
PACs, 5, 42, 63, 104, 105, 127 Rowland, Roy, 90, 91
Pallone, Frank, 91 "rump group" {see also mainstream coali-
paradox of participation, 19 tion), 103, 104, 107
Participation in America, 17
patch-thrus, 66, 88 Safire's Political Dictionary, 1
Payne, L. R, 72, 97, 99, 100, 118 salience of issues, 32, 33, 35, 38, 39, 41,
Pear, Robert, 100, 104, 144 67,74
persuadability Santorum, Rick, 99
constituency targets for, 129 Scarlett, Thomas, 74, 97
of legislators, 47, 56, 78, 83, 89, 104 Schattschneider, E. E., 131
measure of, 79 Schenk, Lynn, 90, 91, 94, 105, 118
petition signing, 21 Schlozman, Kay, 4, 10, 17-21, 23
Pew Center, 116 Second Amendment, 56
Philip Hart Papers, 14 secretary of the Treasury, 1
phone banks, 3, 31, 66, 73, 109 Securities and Exchange Commission, 62
Pickle, J. J., 97, 99, 100 Senate Finance Committee, 1, 43, 78, 80-
Plisner, Marty, 75, 88 82, 89, 101, 103, 107, 108, 126, 145
political demonstrations, 21 Senate Labor and Human Resources, 78,
Pontius, John, 15,16 80-83, 107, 145
Pool, Ithiel de Sola, 23, 127 Sened, Itai, 127
158 Index
Shaw, E. Clay, Jr., 99 town meetings, 108
Sierra Club, 26 traceability, 3, 7, 8, 25, 33, 35, 38, 39, 41,
single-payer health care plan, 82, 100, 144, 42, 48, 53, 55, 56, 57, 60, 62, 65, 66,
145 76, 77, 104
Skocpol, Theda, 73, 85, 144 trade associations, 24, 54, 75, 84, 125
Slattery, Jim, 90, 91, 94-96, 108, 118 Tversky, Amos, 50
Smith, Richard, 32, 52, 127
smoking/tobacco, 7, 54, 65
U.S. League of Savings, 1
Stark, Fortney, 99
Ujifusa, Grant, 80, 89, 90, 97, 101
State of the Union Address, 145
undecideds, 47, 58-61, 68, 70, 78, 83-85,
Stimson, James, 49
90, 91, 103, 106-108, 115, 123, 126-
Stokes, Donald, 18, 127
128
Stone, Peter, 3, 64, 84
unions, 7, 8, 54, 55, 57, 63, 64, 66, 68, 69,
strategic lobbying/tactics, 4 - 6 , 10-13, 2 0 -
82, 84, 85, 110, 112-115, 117-120
22, 28, 30-32, 4 2 - 4 5 , 47, 52-54, 5 7 -
United States Chamber of Commerce, 66,
59, 73, 77, 84, 105, 126, 128, 129,
73
131, 144
University of Michigan, 8, 14
subcommittees, 24, 84, 95
Unsoeld, Jolene, 12
Sullivan, Laurie, 104
Upton, Fred, 91, 94, 118
Sundquist, Don, 99
US News and World Report, 75
survey sampling, 115
Sweeney, John, 26
Swift-Rosenweig, Leslie, 25 Verba, Sidney, 4, 17, 18, 19, 20, 21, 50
Vogel, David, 24
talk radio, 118-120 vote-by-mail, 130
targeting, 22, 26, 42, 43, 46, 47, 49-52, voter turnout, 19-21
66, 67, 105
constituent targets, 6, 10, 12, 13, 20, 26,
Walker, Jack, 8, 10
4 2 - 4 3 , 50, 51, 58, 61, 63-65, 6 9 - 7 1 ,
Wall Street Journal, 7
78, 86-89, 106, 109-121, 126, 127,
Wallop, Malcolm, 102
129
Washington, Craig, 105
legislative targets, 10, 12, 13, 20, 4 6 - 4 8 ,
Washington Post, 7, 24, 25, 75, 88, 96
56, 59-61, 68, 71, 78, 81-85, 89-91,
Washington Representatives, 7
94, 96, 97, 100, 101, 103, 104, 107-
Wayman, Frank, 42, 59, 94, 127
109, 118, 123, 126, 127
Webster decision, 20
targeting software, 62, 66
Webster's New World Dictionary, 1
Tarrance Group, 6, 107
Weisskopf, Michael, 74, 88
Tauzin, Billy, 90, 91, 94, 118
West, Darrell, 77, 84, 88
tax exempt organizations, 26, 27
Wielhouwer, Peter, 49
taxes, 1, 2, 33, 40, 41, 45, 56, 57, 60, 65,
Wittenberg, Elisabeth, 2
67, 70, 144
Wittenberg, Ernest, 2
Taylor, Paul, 1, 2
Wofford, Harris, 72
technological changes in lobbying, 23, 24,
Wolfinger, Raymond, 18, 21
66, 67, 125, 130
Wolpe, Bruce, 2
telecommunications, 7, 54
Woodward, Bob, 57
term limits, 7, 54
worker safety regulations, 7, 54, 69, 70
third-party targets, 64, 65, 69, 87
Thomas, William, 99
Thomson, Margaret, 22 Wright, John, 32, 46, 52, 59, 127
Tierney, John, 4, 10, 23
Times-Mirror Center for the People and the Xerox, 86
Press, 116
Times-Mirror Survey, 6, 13, 17, 108, 116,
117,119-122 Yellowtail, Bill, 27
Toner, Robin, 72, 74, 100, 104 Zaller,John, 18, 127
tort reform, 7, 54, 65 Ziegler, Harmon, 127

(Kenneth M. Goldstein) Interest Groups, Lobbying, (B-Ok - Xyz)

Uploaded by

Document Information

Original Title

Copyright

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

(Kenneth M. Goldstein) Interest Groups, Lobbying, (B-Ok - Xyz)

Uploaded by

Copyright:

Interest Groups, Lobbying, and

Kenneth M. Goldstein is assistant professor of political science at Arizona

© Kenneth M. Goldstein 1999

This book is in copyright. Subject to statutory exception

First published 1999

A catalog record for this book is available from

Library of Congress Cataloging in Publication data

Transferred to digital printing 2003

List of Figures and Tables page viii

Appendix A: Sources and Coding for Survey Data 133

Grass roots 1. The common people.

Grass roots: The ultmate source of power, usually

Congress in opposition to the new law. The effort, orchestrated by the

of legions of financial experts preceded the deletion of caps on credit card

Data and Methodology

Table 1.1. Issue Areas and Frequencies

The forty-one interviews come from a sample of eighty organizations that

association, or corporation that was mentioned as pursuing grass roots tac-

Number in Interviews Completed Mobilization

Plan of the Book

The reigning theories of participation in American

Figure 2.1. Mail to House of Representatives, 1973-1994. Source: Mann and

telegrams. Also, according to my survey of congressional offices, more than

1978 1980 1986 1990 1994

Using the Participation in America data set as a baseline, Verba, Schloz-

The Paradox of Participation

individual probabilities of influencing an election race. In fact, studies indi-

participatory stratification. Their consequences for the representation of publics

Earlier in their study, Verba, Schlozman, and Brady do demonstrate the

"Personal characteristics - resources, perceived rewards, interests, and ben-

Interest Group Lobbying Tactics

Kernell (1993) explained the rise in "going public" by presidents as a

New York Times

Figure 2. 3. Grass roots lobbying mentions in the press, 1977-1994. Source:

Public Affairs Council, a Washington-based group of corporate and trade

Issue Advocacy and Independent Expenditure Campaigns

vertisements - ostensibly asking for citizen action on a legislative issue -

All animals are equal, but some animals are more

Remember, there are only three people in Congress

Most scholarly accounts of interest groups and lobbying have focused on

Influencing Congressional Elections

Challenger Quality Election Outcome

Figure 3.1. Major factors influencing congressional elections.

choice of tactics, interest groups are trying to influence voting behavior in

Incumbent Behavior and Establishing Traceability

television campaign in targeted states and districts seems to have been an

Table 3.1. Strategies for Influencing Congressional Elections

lengers need to be convinced that a race is winnable. Krasno and Green

Influencing Legislative Action

What Is Constituent Intensity?

What Is Legislator Intensity?

Figure 3.2. Legislative decision making.

Providing Information about the Direction and Intensity of Constituency

Framing Issues and Influencing Opinions

Table 3.2. Strategies to Influence Congressional Decisions

Choosing When, Where, Whom, and How to Mobilize

simultaneously recipients and transmitters of strategically targeted elite

Calculating the Influence of a Citizen

Calculating the Effect of a Communication on a Legislator

tion or who are beholden to powerful committee chairs will be minimally

Calculating a Legislator's Probability of Influencing the Process

For groups pursuing a legislative or vote-influencing strategy, the proba-

Objective Strategy Legislative Targets

Calculating and Influencing the Probability That a Citizen Will Contact

Although there is no registration requirement or legal cost for those who

spontaneous activity. Citizens contact their legislators when someone asks